Abstract 159P
Background
The stromal component constitutes as much as 90% of pancreatic cancer specimens, dynamically interacting with the tumor and adapting into a pro-survival environment. This poses a clinical challenge, as biopsies often miss cancer by only sampling stroma. By leveraging AI and image analysis, we aim to extract informative cues from stromal interactions for novel cancer biomarker identification. This approach offers the potential for enhanced diagnostic precision and a deeper understanding of pancreatic cancer biology.
Methods
Anonymized digital scans of pancreatic cancer and chronic pancreatitis were sourced from the Centre Hospitalier de l’Université de Montréal. QuPath 0.4.3 aided slide annotation, with subsequent TIF annotation export. Staining normalization was performed via the Mitkovetta technique in Python. Our process involved deep-learning stromal segmentation, prioritizing >95% stromal tiles using Ilastik. Feature extraction was executed utilizing computer vision techniques (Haralick features), alongside the pre-trained and class-trained ImageNet deep-neural network, VGG16.
Results
Our annotated, normalized, automated, and 95% stroma-probability method generated for the training cohort 9829 cancer and 1638 mass-forming pancreatitis tiles, and 10776 cancer and 1211 pancreatitis tiles for the testing set. The table highlights the performance of the classical computer vision approach (Haralicks features extraction in RGB). Furthermore, transferring the ImageNet VGG-16 pre-trained model to our dataset managed to predict the presence of adjacent cancer at 86.6% accuracy. Table: 159P
Training | Validation | |||
Haralicks features | Cancer (N=9829) vs None (N=1638) | P | Cancer (N=10776) vs None (N=1211) | P |
RGB-F2 | +66% | 7.52 X 10-308 | +8.2% | 1.37 X 10-9 |
RGB-F15 | +54% | 3.28 X 10-272 | +8.5% | 8.57 X 10-11 |
RGB F37 | +9.6% | 2.35 X 10-294 | +2.7% | 5.02 X 10-36 |
Conclusions
We demonstrate that normalized stromal tiles could predict the presence of cancer accurately just by their morphological features at HE staining. This highlights the importance of stroma for diagnostic purposes and can serve as the basis for future studies through multiplex imaging and spatial transcriptomics.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
Vincent Quoc-Huy Trinh.
Funding
Fonds de Recherche Québec Santé.
Disclosure
All authors have declared no conflicts of interest.
Resources from the same session
184P - Neutropenia as a predictive and prognostic factor in nanoliposomal-irinotecan/fluorouracil/leucovorin therapy for pancreatic cancer: Findings from the NAPOLEON-2 study (NN-2301)
Presenter: Yuki Sonoda
Session: Poster Display
Resources:
Abstract
185P - Disease etiology impact on outcomes of hepatocellular carcinoma patients treated with atezolizumab plus bevacizumab: A real-world, multicenter study
Presenter: Silvia Foti
Session: Poster Display
Resources:
Abstract
186P - Efficacy and safety of fruquintinib with nab-paclitaxel in advanced G/GEJ cancer after exposure to immune checkpoint inhibitors: A single-center prospective clinical trial
Presenter: Xiaoting Ma
Session: Poster Display
Resources:
Abstract
187P - Neoadjuvant cadonilimab (PD-1/CTLA-4 bispecific antibody) plus transhepatic arterial infusion chemotherapy (HAIC) for resectable multinodular CNLC Ib/IIa hepatocellular carcinoma (Car-Hero)
Presenter: Yongguang Wei
Session: Poster Display
Resources:
Abstract
188P - Impact of metformin, statin, aspirin and insulin on the prognosis of unresectable HCC patients receiving first-line lenvatinib or atezolizumab plus bevacizumab
Presenter: Margherita Rimini
Session: Poster Display
Resources:
Abstract
189P - Safety run-in results from LEAP-014: First-line lenvatinib (len) plus pembrolizumab (pembro) and chemotherapy (chemo) for metastatic esophageal squamous cell carcinoma (ESCC)
Presenter: Shun Yamamoto
Session: Poster Display
Resources:
Abstract
190P - Perioperative camrelizumab combined with chemotherapy for locally advanced gastric or gastroesophageal junction adenocarcinoma: A single-arm, single-center, phase II clinical trial
Presenter: Jiaxing He
Session: Poster Display
Resources:
Abstract
191P - Predictive value of CXCR6 expression in gastric cancer survival and immune modulation
Presenter: Song-Hee han
Session: Poster Display
Resources:
Abstract
192P - Antiangiogenesis-related adverse events (ARAE) to predict efficacy in patients with advanced gastric cancer (AGC) treated with apatinib + chemotherapy: Results from two prospective studies
Presenter: Rongbo Lin
Session: Poster Display
Resources:
Abstract
193P - Impact of coronavirus disease 2019 on patients with unresectable hepatocellular carcinoma treated with atezolizumab/bevacizumab
Presenter: Hongjae Chon
Session: Poster Display
Resources:
Abstract