Virtual Library
Start Your Search
Rajiv Kumar
Author of
-
+
P33 - Pathology - Immunotherapy Biomarker (ID 101)
- Event: WCLC 2020
- Type: Posters
- Track: Pathology, Molecular Pathology and Diagnostic Biomarkers
- Presentations: 1
- Moderators:
- Coordinates: 1/28/2021, 00:00 - 00:00, ePoster Hall
-
+
P33.18 - The Prevalence of PDL-1 Expression in Lung Cancer: Real-World Experience from a Tertiary Care Oncology Centre (ID 2287)
00:00 - 00:00 | Presenting Author(s): Rajiv Kumar
- Abstract
Introduction
Immune checkpoint inhibitors are the new key players in lung cancer management. Increasing demands for PD-LI(Programmed Death-Ligand 1) testing has been observed recently and its role is emerging as a potential biomarker for immunotherapy. However, there is the paucity of the Indian data on the prevalence of PDL1 expression. This study was undertaken with an objective to evaluate the prevalence of PD-L1 expression in lung cancer patients and analyze its correlation with clinicopathological factors
Methods
All cases of histologically proven lung cancer, diagnosed between January 2017–November 2019, wherein PD-L1 testing was performed were retrieved from electronic medical records. PDL-1 Testing was performed using rabbit Anti-Human PD-L1 monoclonal antibody(Ventana clone SP263) on VENTANA BENCHMARK XT Auto-immmunostainer. Membranous PD-L1 expression of >/ =1% in the tumour cells was regarded as positive. Prevalence of PD-L1 expression and its correlation with clinic-pathological factors were recorded. Further, stratification of the cases was done, in subgroups of: Negative PDL1 (0/<1% PD-L1 expression ), low PDL1 expression(1 – 49%) and high PD-L1 expression (> =50%). The intensity of expression was recorded as weak, moderate & Strong
Results
Out of total 627 clinical request for PDL1 testing in lung cancer during this period, the results could be analyzed in 568 cases(In 35 cases the test could not be performed due to inadequate tumour content and in 23 cases although the test was performed, but could not be interpreted due to technical and interpretative issues). Hence the study cohort comprised of 568 cases. The median age was 56.95 years (range: 29-83years) with male predominance (M:F ratio-2.4:1) and advanced clinical stage III/IV[482(84.8%)]. Smoking history was present in 340 cases(59%). Lung biopsies were the most common type of specimen (n=351,61.7%) followed by metastatic lymph nodes (n=101, 17.8%), effusion cytology cell block (n=36, 6.3%) liver (n=33,5.8%) and others metastatic sites (n=40, 7%). Adenocarcinoma was predominant histological subtype(n=437,76.9% ) followed by squamous cell carcinoma(n=87, 15.3% ), NSCLC(n=13, 2.28%), small cell carcinoma (n=12, 2.11% ) and others(n=19,3.34%).
Overall 57.5 %(n=327/568) of cases revealed PD-L1 positivity(> 1% expression of any intensity) including low PDL1 expression (1-50%) in 37.6 %(n=214)and high PDL1 expression(>50%) was seen in 19.8%(n=113). Almost one-third of positive cases (n=110/327, 33.6%), the PDL1 expression was very low i.e. < 10 % of tumour cells. Most of the cases revealed IHC expression of moderate-intensity (n=160, 48.9%) followed by weak(n=85,26%) and strong intensity in 82 cases(25%). Heterogeneous expression was noted in 74 cases (22.6%). The expression was greater in tumours with squamous histology, smokers and female patients. Further, PDL-1 positivity rates were higher in EGFR mutated tumors70/124(66.7%) as opposed to EGFR wild tumours 224/362(41.9%)
Conclusion
This was the first comprehensive study on the prevalence of PD-L1 expression in Indian lung cancer population demonstrating real-world data. Although the overall positivity rate of PD-L1 expression was 58%, only 19% of cases had high expression i. e in > 50 % tumour cells. Heterogeneity in PD-L1 staining is a crucial factor affecting its evaluation. The findings reinforces the utility of PDL1 testing in Indian patients for segregation of patients for frontline immune checkpoint inhibitors
-
+
P37 - Pathology - Biomarker Testing (ID 107)
- Event: WCLC 2020
- Type: Posters
- Track: Pathology, Molecular Pathology and Diagnostic Biomarkers
- Presentations: 1
- Moderators:
- Coordinates: 1/28/2021, 00:00 - 00:00, ePoster Hall
-
+
P37.14 - Comparative Evaluation of IHC versus FISH for Detection of ROS1 Fusion in Lung Cancer (ID 2135)
00:00 - 00:00 | Presenting Author(s): Rajiv Kumar
- Abstract
Introduction
ROS1 gene rearranged tumours are a rare, yet distinct molecular subgroup of non-small cell lung carcinomas (NSCLC). This study was performed in order to determine the comparative diagnostic accuracy of ROS1 immunohistochemistry(IHC), as opposed to Fluorescence in situ hybridization(FISH) for its ability to detect ROS1 gene rearrangement in NSCLC.
Methods
A validation study designed to assess the diagnostic accuracy of ROS1 IHC using D4D6 (Cell Signaling Technology (CST), Danvers, MA) antibody clone for detection of ROS 1 gene rearrangement as compared to FISH testing using ZytoLight SPEC/ROS1 Dual-Colour Breakapart Probe(ZytoVision, Bremerhaven, Germany) was performed, on an enriched cohort of ROS1 rearranged tumours(including 44 ROS1 rearranged and 166 ROS1 non-rearranged cases). The IHC interpretation was performed by two pathologists independently, who were blinded to the FISH results. Receiver operating characteristics (ROC) curves were used to determine the optimal cut-off value for H-score and proportion of positivity for ROS1 IHC, in order to discriminate between patients with ROS1-rearranged and ROS1- non-rearranged tumours.
Results
Overall ROS1 IHC positivity was observed in 41/210 (19.52%) and 40/210 cases (19.05%) by two different observers, independently, blinded to FISH results, with the almost perfect interobserver agreement [ Kappa value 0.985. (95%CI, 0.95 to 1)]. The immunoexpression was predominantly cytoplasmic and heterogenous with median H-score of 165(range 5-300). Amongst, 44 ROS1gene rearranged cases, 40 cases were positive by ROS1 IHC and amongst, 166 ROS1gene non-rearranged cases, 165 cases were negative by ROS1 IHC. Hence, overall ROS1 IHC had a sensitivity of 90.9%, a specificity of 99.40%, the positive predictive value of 97.56% and negative predictive value of 97.63% considering FISH as the gold standard. A total of 5 discordant cases including one false positive(ROS IHC+/FISH-) and 4 false negatives (ROS IHC-/FISH +) were recorded. ROC curve analysis revealed the optimal cut-off: H-score of ≥2.5 and proportion of positivity ≥2.5%, can best predict the ROS rearrangement using ROS IHC in this study cohort. In 6 cases, of ROS1 rearranged tumours, low IHC expression(i.e only weak intensity staining (1+), positivity in <25% tumour cells and H-score <100) was noted. Although, 3 out of these 6 patients had received the Crizotinib, however, the clinical response was not as promising and developed progressive disease, after a short interval of Crizotinib.
Conclusion
The comparative evaluation in this enriched ROS1 rearranged cohort was unique in term of high specificity(99.40%), and relatively low sensitivity(90.9%) of ROS1 IHC and are somewhat contradictory as compared to previous reports. None of the previously reported cut-off criteria (H score ≥100/150, Staining intensity of ≥2+ in 30% tumour cells) can predict the ROS1 rearrangements with 100%accuracy in our study population and, some cases might be devoid of further molecular testing. Hence, it raises concerns about the utility of D4D6 antibody clone for ROS1 screening. Use of the appropriate cut off interpretative criteria’s can maximize the sensitivity and specificity of ROS1 IHC for predicting the ROS1 rearrangements. The corelation of the level of ROS1 protein expression with response to the Crizotinib needs to be evaluated further.
