LGCmain
Indian Journal of Pathology and Microbiology
Home About us Instructions Submission Subscribe Advertise Contact e-Alerts Ahead Of Print Login 
Users Online: 1598
Print this page  Email this page Bookmark this page Small font sizeDefault font sizeIncrease font size


 
  Table of Contents    
ABSTRACTS  
Year : 2019  |  Volume : 62  |  Issue : 5  |  Page : 20-21
Best Paper Awards Abstracts



Click here for correspondence address and email

Date of Web Publication13-Dec-2019
 

How to cite this article:
. Best Paper Awards Abstracts. Indian J Pathol Microbiol 2019;62, Suppl S1:20-1

How to cite this URL:
. Best Paper Awards Abstracts. Indian J Pathol Microbiol [serial online] 2019 [cited 2020 May 26];62, Suppl S1:20-1. Available from: http://www.ijpmonline.org/text.asp?2019/62/5/20/272755





   BP01: Artificial intelligence in pathology: Recognition of breast cancer from whole slide imaging Top


Parikshit Sanyal

Military Hospital Jalandhar, Jalandhar, Punjab, India

Background and Objective: Histopathologic diagnosis is the gold standard for confirmation of breast carcinoma. However, evaluation of histopathologic slides for recognition of foci of cancer is carried out manually by the pathologist, which is a labour intensive as well as time consuming process. The aim of the present research was to develop a machine learning model for recognition of foci of malignant cells from histopathologic images. Methods: We evaluated a machine learning model, a deep convolutional neural network, built on ResNet 50 architecture, in recognition of foci of breast cancer from histopathological images. The model was trained only on training set with an optimized learning rate. Whole mounted slide images (WSI) dataset prepared by Cruz Roa et al was used for this study. 10094 images of benign foci and 2523 malignant foci were extracted from the WSI images. We split the dataset randomly in training and evaluation set and trained the model with the training set. After completion of training, we evaluated its performance on the evaluation set. Results: After completion of training, the model showed 86.85% sensitivity (recall), 70.43% positive predictive value (precision), 90.94% specificity and 96.53% negative predictive value. Conclusion: We conclude that with further training, ResNet models have the potential to become an effective screening tool for histologic diagnosis of breast cancer.


   BP02: Decoding epigenetics: Histone modification patterns in glioblastoma, a clinicopathological study from a tertiary care centre Top


Archana Balsubramanyam

SRIHER, Chennai, Tamil Nadu, India

Background: Glioblastomas are highly aggressive tumors with fewer than 3% patients being alive at five years after diagnosis. The new landscape of epigenetic research provides hope for the development of newer epigenetics-based drugs; an alternative resource of treatment to combat this otherwise incurable malignancy. Recent data shows that histone modifications play a role in glioma genesis. We studied the expression pattern of H3K18Ac and H4K20triMe in glioblastoma. Methodology: Clinical data from 48 glioblastoma cases over 5years collected and immunohistochemistry (IHC) for H3K18Ac and H4K20triMe performed. Q score was calculated (Q=intensity´% positivity of cells) and data analyzed. Results: Age ranged from 9 years to 79 years with male:female ratio of 1.6:1. Nuclear expression by IHC noted in 60% (28/48) cases for H3K18Ac and 98% cases for (46/48) H4K20trime. Median Q score for H3K18Ac was low (1/12) and H4K20trime was higher (6/12). Survival data for 22 cases could be retrieved. None of the 22 (100%) patients had survived at the end of the study period. Four groups were formed based on low/high Q scores of the two histone modifications. Median follow up was 20 months. Overall survival (OS) collected for 20/48 (41%). Similar median OS noted in both groups stained with H3K18Ac, higher median OS was noted in group 3 with low Q scores for H4K20triMe. No significant correlation noted in the Kaplan meier curve. Conclusion: Further investigations like mRNA profiling are warranted to understand the significance of histone modification signatures in glioblastomas. These patterns are upcoming emerging fields in epigenetic cancer research and could provide a plan to tackle many unanswered questions in glioblastoma for prognostic and treatment purpose.

KEYWORDS: Epigenetics, glioblastoma, histone, survival


   BP03: EZH2 upregulation and H3K27 downregulation predicts aggressive tumor biology of urothelial carcinoma Top


Rasheeda Mohammadali

AIIMS, Bhubaneshwar, Odisha, India

Background: Enhancer zeste homolog 2 (EZH2) is one of the major epigenetic modifiers thought to be involved in transcriptional repression of thousands of target genes through trimethylation of H3K27 (lysine 27 residue of histone H3). Deregulated expression of both EZH2 and H3K27me3 has been implicated in the biological behavior and prognostic outcome of various malignancies. However, there is limited data regarding their role in bladder carcinogenesis. Aims: To assess the expression of EZH2 and H3K27me3 in urinary bladder carcinoma and ascertaining their role in determining the biological behavior. Methods: Ninety-five consecutive urinary bladder carcinoma cases (71.6% high-grade) were included in this study. Immunohistochemical analysis for EZH2 and H3K27me3 was performed in tissue microarray (3 mm and 5 mm). A scoring system including staining intensity and proportion of positively stained neoplastic cells was used for assessment. Based on the cumulative scores the expression pattern was divided into 3 sub-groups: Negative (cumulative score 0), Low-expression (cumulative score 1-4) and High-expression (cumulative score 6-9). Results: EZH2 showed a significant positive correlation with the tumor grade (p=0.001). High expression of EZH2 was identified in approximately 38% cases of high-grade tumor contrary to the low-grade tumor (no expression). The cases with high EZH2 expression showed a significantly high MIB-1 labelling index (Median- 1.0 vs 10.0 vs 30.0; p<0.001). In contrast, negative (27.6% vs 14.3%) and low expression (27.6% vs 7.1%) of H3K27me3 was significantly more common in high-grade cases (p=0.011). Loss/low expression of H3K27me3 was significantly associated with lamina propria invasion (p=0.024) and deep muscle invasion (p=0.01). Interestingly, the invasive fronts within the same tumor showed loss/ weak expression of H3K27me3 and stronger expression of EZH2 as compared to the non-invasive areas. Conclusions: This study establishes an important role of the key epigenetic regulators EZH2 and H3K27me3 in the pathobiology of urothelial carcinomas. Strong expression of EZH2 and weak/loss of expression of H3K27me3 are associated with aggressive histological phenotype. Since, epigenetic alterations are reversible biochemical phenomenon, EZH2 and H3K27me3 may be novel targets for urothelial carcinoma treatment.


   BP04: Immunohistochemical and molecular characterization of endometrial stromal sarcomas Top


Sneha Rao

Basavatarakam Indo-American Cancer Hospital, Hyderabad, Telangana, India

Introduction: Endometrial stromal sarcomas (ESS) are rare and characterized by translocations t(7;17)(p15;q11.2) and t(10;17)(q22;p13) resulting in JAZF1-SUZ12 and YWHAE and FAM22 genes fusions used for defining low grade (LG-ESS) and high grade (HG-ESS) tumors. Aim: The objective of the study was to characterize ESSs using immunohistochemical and molecular markers. Materials and Methods: Patients diagnosed as ESS between January 2014 and December 2018 were included in the study. The slides were reviewed along with a panel of immunohistochemical (IHC) markers, CD 10, cyclin D1, estrogen and progesterone receptor, Ki67 and vimentin and classified according to World Health Organization (2014) criteria into LG-ESS, HG-ESS and undifferentiated uterine sarcoma (UUS). Molecular characterization was performed by fluorescence in situ hybridization using relevant probes. Results: The study included 10 ESS with 5 LG-ESS, 3 HG-ESS and 2 UUS. CD 10 was 100% sensitive and 75% specific for LG-ESS. ER and PR were 100% specific but less sensitive (80%) for LG-ESS. 40% (2/5) LG-ESS demonstrated JAZF1-SUZ12 gene rearrangement. All 3 cases of HG-ESS showed diffuse strong cyclin D1 (>70% nuclei) positivity and were negative for CD10, ER and PR and demonstrated YWHAE gene rearrangement. None of the UUS cases demonstrated this gene rearrangement. Conclusion: JAZF1-SUZ12 and YWHAE-FAM22 gene rearrangement help in accurate characterization of ESS and can be used as diagnostic tools especially when the diagnosis is unclear or difficult. Cyclin D1 can be used as an adjuvant immunomarker for YWHAE gene rearranged HG-ESS.

KEYWORDS: CD 10, Cyclin D1, endometrial stromal sarcoma, ER, JAZF1-SUZ12 gene rearrangement, PR, YWHAE- FAM22 gene rearrangement



Top
Correspondence Address:
Login to access the Email id

Source of Support: None, Conflict of Interest: None


Rights and PermissionsRights and Permissions




 

Top
 
 
  Search
 
    Similar in PUBMED
    Email Alert *
    Add to My List *
* Registration required (free)  


    BP01: Artificial...
    BP02: Decoding e...
    BP03: EZH2 upreg...
    BP04: Immunohist...

 Article Access Statistics
    Viewed335    
    Printed3    
    Emailed0    
    PDF Downloaded90    
    Comments [Add]    

Recommend this journal