4058 Background: Immune checkpoint inhibitors are now standard-of-care for patients with esophageal squamous cell carcinoma (ESCC). The predictive significance of PD-L1 expression in ESCC has been investigated by different antibodies, scoring algorithms and cutoff values in several phase III trials. However, it remains controversial whether PD-L1 is a predictive biomarker for response to PD-1/PD-L1 blockade. Evaluating the concordance of the commercially available PD-L1 assays is essential to understand the discrepancy in the ESCC clinical trials, and helpful to further investigate the predictive value of PD-L1 expression. Methods: 145 archival tumor samples were obtained from 131 patients with ESCC (stage I-IV) at National Taiwan University Hospital. Formalin-fixed, paraffin-embedded archival tumor samples were assessed by three commercially available PD-L1 assays: VENTANA SP263, Dako 22C3 and Dako 28-8 assays. Assays were performed in a College of American Pathologists accredited central laboratory and scored for PD-L1 staining by using multiple metrics including tumor cell score (TC), immune cell score (IC) and combined positive score (CPS) or tumor area percentage (TAP). Analytical concordance was calculated pairwise between assays using the Spearman (ρ) rank correlation coefficient. Classification concordance, including agreement between clinically relevant scoring algorithms, was investigated using positive percent agreement (PPA), negative percent agreement (NPA), and overall percent agreement (OPA) at multiple cutoff values to assess the overlap between populations by using different assays. Results: SP263, 22C3 and 28-8 assays showed good analytical correlation for TC staining (Spearman’s rank correlation coefficient 0.8–0.9). Correlation was lower for IC (Spearman’s rank correlation coefficient 0.59-0.61). There was moderate overlap between populations identified by SP263 and 28-8 or 22C3 based on CPS or TAP algorithm at multiple cutoff values. OPAs for these 3 assays ranged from 68%–88% at various matched algorithms. The SP263 and 22C3 PD-L1 assays appeared relatively more sensitive, assigning a higher proportion of patients as PD-L1 positive or high, compared to 28-8 assay. When using assay-specific clinically relevant algorithm, moderate classification agreement was seen for SP263 versus 22C3 or 28-8. Differences were observed between patient populations with tumor classified as PD-L1 high versus PD-L1 low/negative using CPS≥10, TAP≥10% and TC≥1%. The PPA between 22C3 CPS≥10 and SP263 TAP≥10 or 28-8 CPS≥10 or 28-8 TC≥1% were 75%, 57% and 68% respectively. Conclusions: This study is the first dataset to compare various PD-L1 assays in ESCC. Differences in classification of patients with PD-L1 high versus low/negative using clinically relevant algorithms suggest that caution should be taken when comparing data across the trials.

36 Background: Regorafenib is one of the options of salvage treatment for mCRC. We aimed to explore the prognostic factors of such regorafenib treatment. Methods: We searched the database of National Health Insurance, Taiwan for patients who initiated regorafenib treatment for mCRC from September 1, 2015, to December 31, 2018. Database interrogation of National Death Registry, Taiwan and Taiwan Cancer Registry was conducted for survival data and clinicopathological variables, respectively. Results: A total of 3,643 patients were included in the analysis. The median age was 61.2 years. The primary tumor was left-sided and KRAS mutant in 75.3% and 54.5% of patients, respectively; 28.9% of patients received chemotherapy simultaneously, with irinotecan as the most common companion. The median time to treatment discontinuation (TTD) was 2.3 months (95% confidence interval [CI]: 95% CI: 2.3-2.3), and the median overall survival (OS) was 7.2 months (95% CI: 6.9-7.4). Patients with left-sided primary tumors, compared with patients with right-sided tumors, exhibited significantly longer TTD (median, 2.4 vs. 2.1 months, p < 0.001) and OS (median, 7.6 vs. 6.1 months, p < 0.001). By contrast, TTD and OS of patients under regorafenib treatment were similar regardless the KRAS mutation status ( p = 0.415 and 0.643, respectively). Patients who received chemotherapy along with regorafenib, compared with patients who did not, also exhibited significantly longer TTD (median, 2.6 vs. 2.2 months, p < 0.001) and OS (median, 8.3 vs. 6.7 months, p < 0.001). In multivariate analysis adjusting age, sex, time from colorectal cancer diagnosis, hospital level, KRAS mutation, and initial regorafenib dose, left-sidedness of the primary tumor remained a predictor for better TTD (hazard ratio [HR]: 0.88, p = 0.002) and OS (HR: 0.92, p = 0.048), as well as addition of chemotherapy (for TTD, HR: 0.86, p < 0.001; for OS, HR: 0.76, p < 0.001). Due to the interaction between sidedness and KRAS mutation, we established separate Cox models and found that left-sided primary tumor was an independent predictor for better TTD and OS for KRAS wild-type tumors, but not for KRAS mutant tumors. Conclusions: In this population-wide cohort study, left-sided primary tumor and addition of chemotherapy were associated with better TTD and OS of regorafenib treatment for mCRC.

49 Background: Neoadjuvant chemoradiotherapy (nCRT) followed by surgery is one of indicated strategies for locally advanced ESCC. Pathological complete response (pCR) is a well-known favorable prognostic factor. However, whether pCR to nCRT impacts on the post-progression survival (PPS) is not well described before. Methods: We reviewed medical records of four studies evaluating nCRT for locally advanced ESCC between 2000 and 2015 at National Taiwan University Hospital (NCT00154804, NCT01034189, NCT01034332). CRT comprised twice weekly paclitaxel and cisplatin-based regimens and 40Gy radiotherapy in 20 fractions; esophagectomy with lymph node dissection was performed 6 to 8 weeks after completing nCRT. We analyzed the recurrence patterns and PPS between pCR and non-pCR groups. Results: Among 158 patients (pts) receiving nCRT followed by surgery, 85 pts developed recurrence (recurrence rate [RR], 53.8%) with a median follow-up of 110.2 months (mo). Among pts with recurrence, 19 were from pCR group (RR, 33.3%), and 66 from non-pCR group (RR, 65.3%). The median time to recurrence (TTR) were 12.6 and 8.5 mo for pCR and non-pCR group. The patterns of recurrence were summarized in the table. The median PPS in pCR and non-pCR group were 7.9 and 5.3 mo. In pCR group, the median PPS was similar despite different patterns of recurrence. In contrast, the median PPS of non-pCR group was numerically longer in pts with isolated locoregional recurrence and shortest in those with any distant recurrence (10.0 and 4.8 mo). Pts with recurrence more than 12 mo after surgery exhibited longer PPS than those within 12 mo in both groups. Overall, the median PPS of pts of different recurrence pattern or TTR did not differ significantly between two groups. Conclusions: The recurrence patterns may be different in pCR and non-pCR groups. Patients with early recurrence faced a poor prognosis irrespective of pathologic response. Whether pCR status and recurrence impact on PPS in pts receiving nCRT warrants additional studies.[Table: see text]