A novel method for detection of pancreatic Ductal Adenocarcinoma using explainable machine learning.

BACKGROUND AND OBJECTIVE: Pancreatic Ductal Adenocarcinoma (PDAC) is a form of pancreatic cancer that is one of the primary causes of cancer-related deaths globally, with less than 10 % of the five years survival rate. The prognosis of pancreatic cancer has remained poor in the last four decades, mainly due to the lack of early diagnostic mechanisms. This study proposes a novel method for detecting PDAC using explainable and supervised machine learning from Raman spectroscopic signals. METHODS: An insightful feature set consisting of statistical, peak, and extended empirical mode decomposition features is selected using the support vector machine recursive feature elimination method integrated with a correlation bias reduction. Explicable features successfully identified mutations in Kirsten rat sarcoma viral oncogene homolog (KRAS) and tumor suppressor protein53 (TP53) in the fingerprint region for the first time in the literature. PDAC and normal pancreas are classified using K-nearest neighbor, linear discriminant analysis, and support vector machine classifiers. RESULTS: This study achieved a classification accuracy of 98.5% using a nonlinear support vector machine. Our proposed method reduced test time by 28.5 % and saved 85.6 % memory utilization, which reduces complexity significantly and is more accurate than the state-of-the-art method. The generalization of the proposed method is assessed by fifteen-fold cross-validation, and its performance is evaluated using accuracy, specificity, sensitivity, and receiver operating characteristic curves. CONCLUSIONS: In this study, we proposed a method to detect and define the fingerprint region for PDAC using explainable machine learning. This simple, accurate, and efficient method for PDAC detection in mice could be generalized to examine human pancreatic cancer and provide a basis for precise chemotherapy for early cancer treatment.

A Peptide Meets a Radionuclide to Combat a Rare Tumor.

Background: Neuroendocrine carcinomas (NECs) are rare malignancies with limited treatment options beyond surgery. Peptide receptor radionuclide therapy (PRRT) is a process by which a somatostatin analog (octreotate) is combined with a chelator (DOTA) and a radionuclide (lutetium-177 [177Lu-dotatate]). This therapy targets receptors on neuroendocrine cells, causing internalization of the radionuclide by the tumor cell, which results in cellular damage and apoptosis. Case Report: We describe the clinical and therapeutic course of a 69-year-old male with a metastatic rectal NEC in whom progressive disease was noted after multiple therapies were attempted. After PRRT with 177Lu-dotatate, the patient was asymptomatic and demonstrated a near-complete radiologic response. Conclusion: This case illustrates that treatment with PRRT may improve the outcome of patients with metastatic rectal NEC. Our case highlights the importance of further research into the use of PRRT in patients with a Ki-67 <55% and uptake on somatostatin receptor imaging.

Detection of pancreatic cancer by convolutional-neural-network-assisted spontaneous Raman spectroscopy with critical feature visualization.

Pancreatic cancer is the deadliest cancer type with a five-year survival rate of less than 9%. Detection of tumor margins plays an essential role in the success of surgical resection. However, histopathological assessment is time-consuming, expensive, and labor-intensive. We constructed a lab-designed, hand-held Raman spectroscopic system that could enable intraoperative tissue diagnosis using convolutional neural network (CNN) models to efficiently distinguish between cancerous and normal pancreatic tissue. To our best knowledge, this is the first reported effort to diagnose pancreatic cancer by CNN-aided spontaneous Raman scattering with a lab-developed system designed for intraoperative applications. Classification based on the original one-dimensional (1D) Raman, two-dimensional (2D) Raman images, and the first principal component (PC1) from the principal component analysis on the 2D image, could all achieve high performance: the testing sensitivity, specificity, and accuracy were over 95%, and the area under the curve approached 0.99. Although CNN models often show great success in classification, it has always been challenging to visualize the CNN features in these models, which has never been achieved in the Raman spectroscopy application in cancer diagnosis. By studying individual Raman regions and by extracting and visualizing CNN features from max-pooling layers, we identified critical Raman peaks that could aid in the classification of cancerous and noncancerous tissues. 2D Raman PC1 yielded more critical peaks for pancreatic cancer identification than that of 1D Raman, as the Raman intensity was amplified by 2D Raman PC1. To our best knowledge, the feature visualization was achieved for the first time in the field of CNN-aided spontaneous Raman spectroscopy for cancer diagnosis. Based on these CNN feature peaks and their frequency at specific wavenumbers, pancreatic cancerous tissue was found to contain more biochemical components related to the protein contents (particularly collagen), whereas normal pancreatic tissue was found to contain more lipids and nucleic acid (particularly deoxyribonucleic acid/ribonucleic acid). Overall, the CNN model in combination with Raman spectroscopy could serve as a useful tool for the extraction of key features that can help differentiate pancreatic cancer from a normal pancreas.

Management of Appendiceal Neuroendocrine Tumors: Metastatic Potential of Small Tumors.

BACKGROUND: Appendiceal neuroendocrine tumors (ANETs) are rare neoplasms usually discovered incidentally during appendectomy. ANETs < 2 cm were thought to have no metastatic potential, and this dogma has driven management. Our aim is to evaluate the metastatic potential of ANETs < 2 cm. PATIENTS AND METHODS: A retrospective review was performed in a series of patients with ANETs who presented to our tertiary referral center from 1998 to 2019. Demographics, tumor characteristics, treatment, and clinical outcomes were evaluated. RESULTS: In total, 114 patients were included. Median follow-up was 3.3 years (range, 21 days-15 years). At last follow-up, 34 (30%) patients had positive regional lymph nodes and 20 (18%) patients had metastatic disease. Of the 20 patients with metastatic disease, 11 (55%) had primary ANETs < 2 cm. Patient age > 40 years at diagnosis and ANETs with serosal invasion, lymphovascular invasion, intermediate tumor grade, or positive lymph nodes were features significantly more likely to present with metastatic disease. We found no difference in the rate of lymph node positivity, metastatic disease, or overall survival when patients were stratified by tumor size or type of resection (appendectomy vs. right hemicolectomy). On multivariate analysis, patients with metastatic disease at diagnosis had worse overall survival (HR = 24.4, p = 0.008). CONCLUSIONS: In our cohort, tumor size was not a significant risk factor for metastatic disease or worse outcome as many patients with ANETs < 2 cm developed metastatic disease. Appendectomy alone was sufficient surgical management for most ANETs. Patients with risk factors for metastatic disease, regardless of primary ANET size, should be evaluated thoroughly and counseled for further management and surveillance.

Outcomes of Capecitabine and Temozolomide (CAPTEM) in Advanced Neuroendocrine Neoplasms (NENs).

Capecitabine and temozolomide (CAPTEM) have shown promising results in the treatment of neuroendocrine neoplasms (NEN). The aim of this study was to evaluate the outcome and role for CAPTEM in malignant neuroendocrine neoplasms. Data were obtained from NEN patients who received at least one cycle of CAPTEM between November 2010 and June 2018. The average number of cycles was 9.5. For analysis, 116 patients were included, of which 105 patients (91%) underwent prior treatment. Median progression free survival (PFS) and overall survival (OS) were 13 and 38 months, respectively. Overall response rate (ORR) was 21%. Disease control rate (DCR) was 73% in all patients. PFS, median OS, ORR, and DCR for pancreatic NENs (pNEN) vs. non-pNEN was 29 vs. 11 months, 35 vs. 38 months, 38% vs. 9%, and 77% vs. 71%, respectively. Patients with pNEN had a 50% lower hazard of disease progression compared to those with non-pNEN (adjusted Hazard Ratio: 0.498, p = 0.0100). A significant difference in PFS was found between Ki-67 < 3%, Ki-67 3-20%, Ki-67 > 20-54%, and Ki-67 ≥ 55% (29 vs. 12 vs. 7 vs. 5 months; p = 0.0287). Adverse events occurred in 74 patients (64%). Our results indicate that CAPTEM is associated with encouraging PFS, OS, and ORR data in patients with NENs.

Plasma Pancreastatin Predicts the Outcome of Surgical Cytoreduction in Neuroendocrine Tumors of the Small Bowel.

OBJECTIVES: Elevated pancreastatin (PST) levels have been shown to be associated with poor prognosis in small bowel neuroendocrine tumors (NETs). We hypothesized that plasma PST levels that remain elevated following surgical cytoreduction portend a poor prognosis in well-differentiated small bowel NETs. METHODS: Patients diagnosed with small bowel NETs who underwent surgical cytoreduction at our institution were identified. Demographics, histopathologic characteristics, and biochemical data were collected. Only patients who had serial preoperative PST (PreopPST) and postoperative PST (PostopPST) levels were included in this study. Patients were sorted into groups by PST level to assess their response to surgical cytoreduction (group 1, PreopPST/PostopPST normal; group 2, PreopPST elevated/PostopPST normal; group 3, PreopPST/PostopPST elevated). Survival rates were calculated from the date of surgery. RESULTS: PreopPST and PostopPST levels were collected from 300 patients. Patients in groups 1 (n = 74) and 2 (n = 81) had a significant survival advantage compared with patients in group 3 (n = 145) (P < 0.0001). Kaplan-Meier 5- and 10-year survival rates were as follows: group 1: 93% and 82%; group 2: 91% and 65%; and group 3: 58% and 34%, respectively. CONCLUSIONS: Serial monitoring of plasma PST is useful in predicting long-term survival following surgical cytoreduction and can be helpful to identify patients who have a poor prognosis.

Current Treatment Options in Gastroenteropancreatic Neuroendocrine Carcinoma.

Poorly differentiated gastroenteropancreatic neuroendocrine carcinomas (GEPNECs) are a rare neoplasm with a bleak prognosis. Currently there are little prospective data available for optimal treatment. This review discusses the current available regimens and the future direction for the treatment of GEPNECs. Treatment plans for GEPNECs are often adapted from those devised for small cell lung cancer; however, differences in these malignancies exist, and GEPNECs require their own treatment paradigms. As such, current first-line treatment for GEPNECs is platinum-based chemotherapy with etoposide. Studies show that response rate and overall survival remain comparable between cisplatin and carboplatin versus etoposide and irinotecan; however, prognosis remains poor, and more efficacious therapy is needed to treat this malignancy. Additional first-line and second-line treatment options beyond platinum-based chemotherapy have also been investigated and may offer further treatment options, but again with suboptimal outcomes. Recent U.S. Food and Drug Administration approval of peptide receptor radionuclide therapy in low- and intermediate-grade neuroendocrine tumors may open the door for further research in its usefulness in GEPNECs. Additionally, the availability of checkpoint inhibitors lends promise to the treatment of GEPNECs. This review highlights the lack of large, prospective studies that focus on the treatment of GEPNECs. There is a need for randomized control trials to elucidate optimal treatment regimens specific to this malignancy. IMPLICATIONS FOR PRACTICE: There are limited data available for the treatment of poorly differentiated gastroenteropancreatic neuroendocrine carcinomas (GEPNECs) because of the rarity of this malignancy. Much of the treatment regimens used in practice today come from research in small cell lung cancer. Given the poor prognosis of GEPNECs, it is necessary to have treatment paradigms specific to this malignancy. The aim of this literature review is to summarize the available first- and second-line GEPNEC therapy, outline future treatments, and highlight the vast gap in the literature.

Plasma Neurokinin A Levels Predict Survival in Well-Differentiated Neuroendocrine Tumors of the Small Bowel.

OBJECTIVES: Elevated neurokinin A (NKA) levels are associated with poor prognosis in patients with small bowel neuroendocrine tumors. We hypothesized that patients with NKA levels that remain elevated despite treatment with surgical cytoreduction have a poor prognosis. METHODS: Patients diagnosed with small bowel neuroendocrine tumors who underwent surgical cytoreduction at our institution were identified. Demographics, histopathologic characteristics, and biochemical data were collected. Patients were grouped by the trend of their NKA levels (group 1, continuously normal; group 2, transiently elevated but normalized after therapy; group 3, remained elevated despite therapy). Survival rates were calculated from the date of the patient's first NKA level. RESULTS: Serial NKA values after surgical cytoreduction were monitored in 267 patients. Kaplan-Meier 2-year, 5-year, and 10-year survival rates were as follows: group 1 (n = 157), 97%, 89%, and 62%; group 2 (n = 78), 99%, 90%, and 78%; and group 3 (n = 32), 88%, 69%, and 0%. Survival rates were statistically significant between groups 1 and 3 and between groups 2 and 3 (P < 0.01). CONCLUSIONS: Serial monitoring of plasma NKA levels is useful in identifying patients who have a poor prognosis. Elevated NKA levels can indicate the need for immediate therapeutic intervention.

Aggressive Surgical Approach to the Management of Neuroendocrine Tumors: A Report of 1,000 Surgical Cytoreductions by a Single Institution.

BACKGROUND: Neuroendocrine tumors (NETs) are rare neoplasms. Our group has treated more than 2,000 NET patients and has performed more than 1,000 surgical cytoreductive procedures. STUDY DESIGN: Records of 834 NET patients who underwent surgical cytoreduction at our institution were reviewed. Demographic information, intraoperative findings, extent of disease, complications, and survival rates were calculated. RESULTS: Eight hundred patients underwent 1,001 cytoreductive operations. Sixty-five percent had small bowel primaries. One hundred and thirty-eight patients presented with an unknown primary site, which was localized intraoperatively in 89% of these cases. The intraoperative complication rate was 9%. The incidence of intraoperative carcinoid crisis was 1%. Mean ± SD operative time was 368 ± 146 minutes. Mean ± SD hospital stay was 9 ± 10 days. Minor postoperative complications occurred after 43% of procedures and major postoperative complications were noted after 19% of procedures. The 30-day postoperative mortality rate was 2%. Median overall survival (OS) for patients with pancreatic NETs was 124 months. The 5-, 10-, and 20-year OS rates for patients with pancreatic NETs were 67%, 51%, and 36%, respectively. The life expectancy difference (between OS and actuarial survival) after surgical cytoreduction for patients with pancreatic NETs was 16.6 years. Median OS for patients with small bowel NETs was 161 months. The 5-, 10-, and 20-year OS rates for patients with small bowel NETs were 84%, 67% and 31%, respectively. The life expectancy difference after surgical cytoreduction for patients with small bowel NETs was 11.7 years. CONCLUSIONS: Surgical cytoreduction in NET patients has low morbidity and mortality rates and results in prolonged survival. We believe that surgical cytoreduction should play a major role in the care of patients with NETs.

Elevated Plasma Pancreastatin, but Not Chromogranin A, Predicts Survival in Neuroendocrine Tumors of the Duodenum.

BACKGROUND: Neuroendocrine tumors (NETs) of the duodenum are rare, heterogeneous, and often indolent neoplasms. We hypothesized that elevated pancreastatin levels are an indicator of a poor prognosis in well-differentiated duodenal NETs. STUDY DESIGN: Data from patients diagnosed with a primary duodenal NET were analyzed. Patients that underwent esophogogastroduodenoscopy, endoscopic ultrasound, or exploratory surgery to localize their neoplasm and whose tumors were confirmed histologically were included. RESULTS: Eighty-four patients were diagnosed with duodenal NETs from January 1991 to January 2014. Seventy-five percent and 21% of patients had their tumor localized by esophogogastroduodenoscopy and endoscopic ultrasound, respectively. The remaining 4% were localized during exploratory surgery. The 5-year Kaplan-Meier survival rate for the entire cohort (N = 84) was 80%. Survival sorted by normal vs abnormal pancreastatin level was statistically significant (p < 0.0001). Five-year survival rates were 94% and 37% for normal and abnormal pancreastatin, respectively. In contrast, survival sorted by normal vs abnormal plasma chromogranin A level was not statistically significant (p = 0.24). CONCLUSIONS: Patients with primary duodenal NETs have high 5-year survival rates. Serial monitoring of plasma pancreastatin levels can identify patients who have a poor prognosis.