Abstract 2096: A deep learning system to predict disease-specific survival in stage II and stage III colorectal cancer

2020 ◽  
Author(s):  
Ellery Wulczyn ◽  
David F. Steiner ◽  
Melissa Moran ◽  
Markus Plass ◽  
Robert Reihs ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Deep Learning ◽  
Learning System ◽  
Stage Ii ◽  
Stage Iii ◽  
Disease Specific Survival ◽  
Disease Specific

scholarly journals Interpretable survival prediction for colorectal cancer using deep learning

2021 ◽  
Vol 4 (1) ◽  
Author(s):  
Ellery Wulczyn ◽  
David F. Steiner ◽  
Melissa Moran ◽  
Markus Plass ◽  
Robert Reihs ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Deep Learning ◽  
Learning System ◽  
Survival Prediction ◽  
Disease Specific Survival ◽  
Clinicopathologic Features ◽  
Visual Appearance ◽  
Prognostic Features ◽  
Poorly Differentiated ◽  
Disease Specific

AbstractDeriving interpretable prognostic features from deep-learning-based prognostic histopathology models remains a challenge. In this study, we developed a deep learning system (DLS) for predicting disease-specific survival for stage II and III colorectal cancer using 3652 cases (27,300 slides). When evaluated on two validation datasets containing 1239 cases (9340 slides) and 738 cases (7140 slides), respectively, the DLS achieved a 5-year disease-specific survival AUC of 0.70 (95% CI: 0.66–0.73) and 0.69 (95% CI: 0.64–0.72), and added significant predictive value to a set of nine clinicopathologic features. To interpret the DLS, we explored the ability of different human-interpretable features to explain the variance in DLS scores. We observed that clinicopathologic features such as T-category, N-category, and grade explained a small fraction of the variance in DLS scores (R2 = 18% in both validation sets). Next, we generated human-interpretable histologic features by clustering embeddings from a deep-learning-based image-similarity model and showed that they explained the majority of the variance (R2 of 73–80%). Furthermore, the clustering-derived feature most strongly associated with high DLS scores was also highly prognostic in isolation. With a distinct visual appearance (poorly differentiated tumor cell clusters adjacent to adipose tissue), this feature was identified by annotators with 87.0–95.5% accuracy. Our approach can be used to explain predictions from a prognostic deep learning model and uncover potentially-novel prognostic features that can be reliably identified by people for future validation studies.

scholarly journals Prognostic Impact of the Number of Examined Lymph Nodes in Stage II Colorectal Adenocarcinoma: A Retrospective Study

2020 ◽  
Vol 2020 ◽  
pp. 1-8
Author(s):  
Purun Lei ◽  
Ying Ruan ◽  
Jianpei Liu ◽  
Qixian Zhang ◽  
Xiao Tang ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Lymph Node ◽  
Lymph Nodes ◽  
Stage Ii ◽  
Lymph Node Status ◽  
Stage Migration ◽  
Prognostic Impact ◽  
Disease Specific Survival ◽  
Stage Ii Colorectal Cancer ◽  
Disease Specific

Background. Evaluation of lymph node status is critical in colorectal carcinoma (CRC) treatment. However, as patients with node involvement may be incorrectly classified into earlier stages if the examined lymph node (ELN) number is too small and escape adjuvant therapy, especially for stage II CRC. The aims of this study were to assess the impact of the ELN on the survival of patients with stage II colorectal cancer and to determine the optimal number. Methods. Data from the US Surveillance, Epidemiology, and End Results (SEER) database on stage II resected CRC (1988-2013) were extracted for mathematical modeling as ELN was available since 1988. Relationship between ELN count and stage migration and disease-specific survival was analyzed by using multivariable models. The series of the mean positive LNs, odds ratios (ORs), and hazard ratios (HRs) were fitted with a LOWESS (Locally Weighted Scatterplot Smoothing) smoother, and the structural break points were determined by the Chow test. An independent cohort of cases from 2014 was retrieved for validation in 5-year disease-specific survival (DSS). Results. An increased ELN count was associated with a higher possibility of metastasis LN detection (OR 1.010, CI 1.009-1.011, p<0.001) and better DSS in LN negative patients (OR 0.976, CI 0.975-0.977, p<0.001). The cut-off point analysis showed a threshold ELN count of 21 nodes (HR 0.692, CI 0.667-0.719, p<0.001) and was validated with significantly better DSS in the SEER 2009 cohort CRC (OR 0.657, CI 0.522-0.827, p<0.001). The cut-off value of the ELN count in site-specific surgeries was analyzed as 20 nodes in the right hemicolectomy (HR 0.674, CI 0.638-0.713, p<0.001), 19 nodes in left hemicolectomy (HR 0.691, CI 0.639-0.749, p<0.001), and 20 nodes in rectal resection patients (HR 0.671, CI 0.604-0.746, p<0.001), respectively. Conclusions. A higher number of ELNs are associated with more-accurate node staging and better prognosis in stage II CRCs. We recommend that at least 21 lymph nodes be examined for accurate diagnosis of stage II colorectal cancer.

scholarly journals High Tumour Cannabinoid CB1 Receptor Immunoreactivity Negatively Impacts Disease-Specific Survival in Stage II Microsatellite Stable Colorectal Cancer

PLoS ONE ◽  
2011 ◽  
Vol 6 (8) ◽  
pp. e23003 ◽  
Author(s):  
Sofia B. Gustafsson ◽  
Richard Palmqvist ◽  
Maria L. Henriksson ◽  
Anna M. Dahlin ◽  
Sofia Edin ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Cb1 Receptor ◽  
Stage Ii ◽  
Disease Specific Survival ◽  
Cannabinoid Cb1 Receptor ◽  
Disease Specific

Clinical Prognostic Factors in Malignant Parotid Gland Tumors

2005 ◽  
Vol 133 (5) ◽  
pp. 702-708 ◽  
Author(s):  
Roberto A. Lima ◽  
Marcos R. Tavares ◽  
Fernando L. Dias ◽  
Jacob Kligerman ◽  
Marilene F. Nascimento ◽  
...  
Keyword(s):  
Lymph Nodes ◽  
Stage Iv ◽  
Stage I ◽  
Stage Ii ◽  
Stage Iii ◽  
Rank Test ◽  
Disease Specific Survival ◽  
Duct Carcinoma ◽  
Neck Lymph Node ◽  
Disease Specific

OBJECTIVE: To analyze the factors in parotid malignant epithelial tumors influencing recurrences and disease-specific survival. METHODS: We retrospectively reviewed the files of 150 patients treated at our institution, from 1974 to 1998. Twenty-four patients were not treated by surgery and were excluded from this study. The remaining 126 patients were treated with surgery and 74 patients had postoperative radiotherapy. Thirty-three patients were treated with parotidectomy plus neck dissection. Neck lymph node metastasis was found in 22 patients, 5 patients had occult neck metastasis, and 4 periparotid lymph nodes metastasis. The mean age was 49 years old. According to the UICC/1997 TNM Classification, 49 patients were stage I, 27 stage II, 22 stage III, and 28 stage IV. The influence of selected factors on the 10 year disease-specific survival was analyzed using the Kaplan-Meier actuarial method and the log-rank test. RESULTS: Forty patients had mucoepidermoid carcinoma, 18 patients adenocarcinoma NOS, 18 patients acinic cell carcinoma, 15 patients adenoid cystic carcinoma, 11 patients malignant mixed tumor, 11 patients salivary duct carcinoma, and 13 patients other pathology. Twenty-five patients had recurrences: 17 had local recurrences, 4 patients had neck recurrences, and 4 were loco-regional recurrences. Five factors influenced negatively the prognosis: 1) T stage (p.00001), 2) grade (p.00001), 3) + lymph nodes (p.0007), 4) facial nerve dysfunction (p.0001), and 5) age (p.004). Patients with high-grade tumors and high-stage tumors had the worst prognosis according to the multivariate analysis. The 10-year disease-specific survival was 97% for stage I, 81% for stage II, 56% for stage III, and 20% for stage IV. CONCLUSION: The grade of the tumor and stage were the most important prognostic factor.

The efficacy of adjuvant chemotherapy for resected high-risk stage II and stage III colorectal cancer in frail patients

2021 ◽  
Author(s):  
Kosuke Mima ◽  
Nobutomo Miyanari ◽  
Keisuke Kosumi ◽  
Takuya Tajiri ◽  
Kosuke Kanemitsu ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
High Risk ◽  
Adjuvant Chemotherapy ◽  
Stage Ii ◽  
Stage Iii ◽  
Frail Patients

scholarly journals Transcriptome of Tumor-Infiltrating T Cells in Colorectal Cancer Patients Uncovered a Unique Gene Signature in CD4+ T Cells Associated with Poor Disease-Specific Survival

Vaccines ◽  
2021 ◽  
Vol 9 (4) ◽  
pp. 334
Author(s):  
Salman M. Toor ◽  
Varun Sasidharan Nair ◽  
Reem Saleh ◽  
Rowaida Z. Taha ◽  
Khaled Murshed ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
T Cells ◽  
T Cell Activation ◽  
Cd4 T Cells ◽  
Poor Prognosis ◽  
Gene Signature ◽  
The Cancer Genome Atlas ◽  
Disease Specific Survival ◽  
Unique Gene ◽  
Disease Specific

Colorectal cancer (CRC) is influenced by infiltration of immune cell populations in the tumor microenvironment. While elevated levels of cytotoxic T cells are associated with improved prognosis, limited studies have reported associations between CD4+ T cells and disease outcomes. We recently performed transcriptomic profiling and comparative analyses of sorted CD4+ and CD8+ tumor-infiltrating lymphocytes (TILs) from bulk tumors of CRC patients with varying disease stages. In this study, we compared the transcriptomes of CD4+ with CD8+ TILs. Functional annotation pathway analyses revealed the downregulation of inflammatory response-related genes, while T cell activation and angiogenesis-related genes were upregulated in CD4+ TILs. The top 200 deregulated genes in CD4+ TILs were aligned with the cancer genome atlas (TCGA) CRC dataset to identify a unique gene signature associated with poor prognosis. Moreover, 69 upregulated and 20 downregulated genes showed similar trends of up/downregulation in the TCGA dataset and were used to calculate “poor prognosis score” (ppScore), which was significantly associated with disease-specific survival. High ppScore patients showed lower expression of Treg-, Th1-, and Th17-related genes, and higher expression of Th2-related genes. Our data highlight the significance of T cells within the TME and identify a unique candidate prognostic gene signature for CD4+ TILs in CRC patients.

PP 3 S-100B concentrations predict disease specific survival in AJCC Stage III melanoma patients

2011 ◽  
Vol 47 ◽  
pp. S21
Author(s):  
S. Kruijff ◽  
E. Bastiaannet ◽  
M. Speijers ◽  
I. Kema ◽  
R. van Ginkel ◽  
...  
Keyword(s):  
Stage Iii ◽  
Disease Specific Survival ◽  
Ajcc Stage ◽  
Melanoma Patients ◽  
Stage Iii Melanoma ◽  
Disease Specific

scholarly journals A Cross Sectional Study on Loco-Regional Recurrences and Overall Survival Rate among Patients with Stage II (T2N0M0) Glottis Carcinoma Treated with Chemoradiation and Surgery

2021 ◽  
Vol 8 (04) ◽  
pp. 219-223
Author(s):  
Niharika Darasani
Keyword(s):  
Overall Survival ◽  
Survival Rate ◽  
Squamous Cell ◽  
Stage Ii ◽  
Overall Survival Rate ◽  
Disease Specific Survival ◽  
Free Survival ◽  
Tertiary Teaching ◽  
Voice Preservation ◽  
Disease Specific

BACKGROUND Single modality treatment for stage I and stage II squamous cell carcinomas of glottis region gave excellent results. Since a long time these are treated either with definitive radiation therapy or surgical excision with endoscopes. There was not much difference with regard to voice preservation, local recurrence and disease-free survival period. Our aim was to study the clinical presentation and management protocol of glottis carcinoma in a tertiary hospital and observe the final outcome of stage II (T2N0M0) glottis carcinoma and specific factor for survival in patients treated with surgery, radiotherapy and concurrent chemoradiation. METHODS 43 patients of glottis carcinoma stage II (T2N0M0) attending a tertiary teaching hospital between May 2015 and April 2017 were included in the study. Demography and smoking status of subjects were recorded. Staging of the disease was according to American Joint Committee on Cancer (AJCC) Staging System 7th edition. Paraglottic space infiltration was taken as a criteria to upgrade the staging. The overall survival rate, recurrence free survival, disease specific survival rate and laryngeal function preservation rate were calculated. RESULTS Out of 43 patients, males were 90.69 % and 09.30 % were females. Male to female ratio was 10.57 : 1. Mean age was 58.62 ± 2.35 years. 67.44 % were current smokers, 27.90 % were former smokers and 02.32 % were non-smokers. The overall survival scores and disease specific survival was 100 % with 11.62 % locoregional recurrences. The voice preservation was 86.04 %. Radiotherapy was used in 72.09 %, chemoradiation in 18.60 % patients and 11.62 % patients underwent surgery. 11.62 % patients presented with locoregional recurrence during 24 months of follow up. 02.32 % patients had to undergo tracheostomy. CONCLUSIONS The overall survival scores and disease specific survival were 100 % with 11.62 % loco-regional recurrence. Voice preservation was 86.04 %. Proactive prevention rather than escalation of treatment protocol gives better prognosis. KEYWORDS Glottis, Larynx, Supra Glottis, Sub Glottis, Squamous Cell Carcinoma, Chemo Radiation and Trans Oral Laryngeal Surgeries

Sign in / Sign up

Export Citation Format

Share Document