الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 77 |  |  |
| | | from | 77 | | |
Abstract
Lung cancer is the leading cause of worldwide cancer-related deaths while, in Egypt, it is the fourth. NSCLC constitutes 85% of primary lung cancers. On the other hand, HNSCC is the seventh top cause of global cancer-related deaths. In the management of both advanced NSCLC and HNSCC, the role of immunotherapy has been established though this has come with several inconsistencies in treatment outcomes. There is need to harness the predictive power of PD-L1 as a marker of response to immunotherapy. This was a quasi-experimental clinical trial investigating inter-tumor heterogeneity of PD-L1 expression between primary tumors and their respective nodal metastasis in both advanced NSCLC and advanced HNSCC among patients in Alexandria, Egypt.
In the NSCLC group, analysis on 15 patients revealed low PD-L1 expression in those who received neo-adjuvant chemotherapy (NACT). Discrepancy in PD-L1 positivity rate among paired primary-nodal tumors was seen at the 50% TPS cut off. Cohen’s kappa coefficient test highlighted statistically significant differences in agreement between the primary and nodal tumor sites at the 1% and 50% TPS cut off points. Histology-specific analysis showed that ADC had a higher concordance rate than SCC (90% versus 66.7%) at the TPS cut off of 1%. Analysis based on clinically relevant PD-L1 clusters indicated that, in 26.67% of the patients, results from tests conducted on primary tumors were shifted to a different cluster when compared with results from PD-L1 tests ran on lymph node specimen. This could potentially affect outcomes following anti-PDL1 therapy. At the TPS cut off of 1%, no association was established between PD-L1 expression and clinic-pathological features in both primary tumors and nodal subgroups; however, presence of LVI was statistically significantly higher in the PD-L1 positive patients in both subgroups.
In the HNSCC group, analysis on 13 patients revealed low positive PD-L1 expression (CPS<10) in both primary and nodal resection specimens among patients who received NACT. At the low cut off of 1%, high PD-L1 positivity rates were reported in both primary and lymph node tumors using the CPS scoring system, whose Negative Predictive Value (NPV) is in question. PD-L1 positivity rate was actually higher in primary tumors than lymph node tumors at both CPS 1% and 20% cut off points. Analysis of agreement between paired biopsies demonstrated a 92.31% Overall Percent Agreement (OPA) at the CPS cut off of 1%, though the kappa value could not be established. However, based on Cohen’s kappa coefficient test, there was only minimal agreement at the CPS cut off of 20%. There was no association between PD-L1 expression and clinic-pathological features at the CPS cut off of 20% in both primary and lymph node subgroups. Survival analysis in the HNSCC group assessed both PFS and OS; it showed no statistically significant survival benefit from PD-L1 expression in both primary and nodal subgroups.
In light of these findings, the study suggests cautious use of PD-L1 as a marker for predicting response to immunotherapy in both NSCLC and HNSCC. In a bid to improve PD-L1’s predictive power, the study emphasizes special consideration in the choice of histological specimen type, tissue sampling sites, as well as the scoring system for PD-L1.