Search

9036 Background: Despite immune-checkpoint inhibitors (ICI) achieving high response rates in patients (pts) with advanced non-small cell lung cancer (NSCLC), immune-related adverse events (irAE) remain an important therapeutic hurdle. The gut microbiome represents a novel prognostic marker of ICI response and early clinical evidence suggests that the microbiome may also regulate the development of irAE. We aimed to assess the association between irAE and the gut microbiome composition in advanced NSCLC pts amenable to ICI. Methods: In this study, we enrolled 464 pts from two independent cohorts (Montreal and Tokyo) with advanced NSCLC treated with ICI monotherapy or combination with chemotherapy. Fecal samples were collected prior to ICI initiation. Metagenomics microbiome profiling and 16S rRNA microbiome sequencing were performed for 81 pts and 133 pts respectively. The irAE were classified as CTCEA grade 0-1 (G0-1 irAE) and grade > 2 (G2-5 irAE). Microbiome composition of both groups was represented using alpha diversity, beta diversity indexes and LEfSe relative abundances. Results: Median follow-up was 28.3 months (mos) and the incidence of G2-5 irAE was 25.1% in the Montreal cohort. For the Tokyo cohort, the median follow-up was 19.6 mos and 30.8% pts developed G2-5 irAE. First, in both cohorts, overall survival (OS) and progression-free survival (PFS) were significantly improved for pts with G2-5 irAE compared to G0-1 irAE (OS: HR: 1.71, p = 0.008 and PFS: HR: 1.65, p < 0.001) and (OS: HR: 2.34, p = 0.001, and PFS: HR = 2.34, p = 0.006) respectively. Second, metagenomics analysis of 81 pts demonstrated a numerical decrease of the alpha diversity in terms of observed species in the G2-5 irAE. Dorea sp CAG 31, Anaerosporobacter mobilis, Butyricicococcus, and Enterococcus faecium were overrepresented in pts with G2-5 irAE. Conversely, Gordonibacter was increased in G0-1 irAE. Next, 16S rRNA profiling from the Tokyo cohort revealed an enrichment of Dorea, Butyricicococcus, and Eubacterium ventriosum in G2-5 irAE . Finally, we observed an enrichment in Dorea and Butyricicococcus in the pts with PFS > 6 months, as observed in the G2-5 irAE group. Conclusions: IrAE correlated with clinical outcome and microbiome composition in two independent cohorts of pts with advanced NSCLC. These results prompt further research on the potential mechanism underling the role of the microbiome in the development of irAE.

9050 Background: Host-related inflammatory biomarkers and gut microbiome are two major prognostic factors for non-small cell lung cancer (NSCLC) patients (pts) treated with immune checkpoint inhibitors (ICI). In this study, we aimed to assess an association between the Lung Immune Prognostic Index (LIPI) and the microbiome composition on ICI outcomes in two independent cohorts of NSCLC pts. Methods: We included 205 patients with advanced NSCLC treated with ICI (monotherapy or in combination with chemotherapy) from two independent cohorts. Metagenomics microbiome profiling was performed on the Canadian discovery cohort of 72 pts while 16S rRNA microbiome sequencing was used in the Japanese validation cohort of 133 pts. The LIPI score was calculated using the dNLR (neutrophils/[leucocytes-neutrophils]) and lactate deshydrogenase (LDH). Pts were classified as Good (G: 0 high factor), Intermediate (I: 1 high factor) and Poor (P: 2 high factors). Median overall survival (OS) was estimated using the Kaplan-Meier method. Microbiome diversity indexes and bacterial relative abundances were compared according to LIPI groups. Results: Among the 72 pts included in the discovery cohort, the median follow-up of 20.6 months (mos). The LIPI was distributed as follows: G (n = 31, 43.1%), I (31, 43.1%), P (10, 13.8%) and baseline characteristics were well balanced between the 3 groups. When segregating pts according to LIPI, the OS was 25.6 mo, 19.8 mo, and 5.7 mo in the G, I (HR: 1.71, 95%CI: 0.80-3.65) and P (HR: 3.97, 95%CI: 1.60-9.82) groups, respectively (p = 0.003). The microbiome alpha diversity was lower in the P group compared with the G group (p = 0.03), and there was a trend towards different microbiome composition in beta diversity (p = 0.055) between both groups. Pts in the G group had a favorable microbiome (enriched in Ruminococcus and Anaerostipes), while pts from the P group had an unfavorable microbiome (enriched in Enterobacteriaceae and Clostridium symbiosum and lavalense). Next, in the validation cohort of 133 pts, LIPI was distributed as follows: G (n = 62, 46.6%), I (51, 38.3%), P (20, 15%). Pts with G LIPI had not reached their median OS compared to pts in the I [15.7 mo HR: 1.60 (0.88-2.94)] and P [8.8 mo groups, HR: 2.02 (0.98-4.19)], p = 0.03. Similar to the discovery cohort, at the genus level, G LIPI group had enrichment of Ruminococcus as well as Anaerostipes compared with pts in the P and I groups with an overrepresentation of Hungatella. Conclusions: Host-related inflammatory biomarkers, represented by the LIPI, seemed to be associated with microbiome and ICI outcomes in pts treated with NSCLC. This observation was validated in an external validation cohort. This link could be in relation to the presence of proinflammatory bacteria in pts with poor LIPI.

9061 Background: There are two types of pembrolizumab-containing strategies for patients with non-small cell lung cancer (NSCLC) exhibiting a high expression level of programmed death-ligand 1 (PD-L1) (tumor proportion score [TPS] ≥50%): the early combination of pembrolizumab plus chemotherapy, and chemotherapy after pembrolizumab failure. Which strategy is superior, however, remains unclear. Comparing progression-free survival (PFS) or progression after the next therapy line (PFS2) in previous clinical trials has not allowed any conclusions regarding superiority to be made. Instead, the time to failure of strategy (TFS), which represents the time until disease exacerbation when the same number of drugs has been used, should be used to compare the two strategies. Methods: We consecutively reviewed the efficacy and safety of first-line, pembrolizumab-containing regimens administered between December 2017 and November 2020. We divided the patients who received pembrolizumab as a first-line treatment into two groups according to whether they received chemotherapy: a pembrolizumab plus chemotherapy group (combo group), and a pembrolizumab monotherapy group (mono group). TFS was defined as the PFS in the combo group and the PFS2 in the mono group. We used the propensity score matching (PSM) method to reduce the bias. Results: Of the 964 patients with advanced NSCLC who underwent first-line treatment, 126 with a PD-L1 TPS ≥50% were eligible for inclusion in this analysis (89 in mono group, 37 in combo group). PSM matched 36 people from each of the two groups. The median follow-up period was 16.2 months (range, 0.1-34.3 months). The patient backgrounds were similar. The overall response rate (ORR) of the combo group was higher than that of the mono group (69.4% vs. 50.0%). The median PFS (mPFS) in the combo group was longer (11.4 months vs. 6.0 months). However, the median TFS (mTFS) of the two groups was almost the same (11.4 months vs. 11.7 months). At the time of the analysis, the median overall survival had not been reached. The frequency of all immune-related serious adverse events (irSAE) was similar, however, that of all SAE and AE leading to treatment discontinuation were larger in the combo group. Conclusions: The ORR of the combo group was higher than that of the mono group; however, the TFS was similar. We suggest that pembrolizumab plus chemotherapy, which can increase toxicity, might be of value in patients, producing a clinically meaningful increase in the ORR.[Table: see text]

TPS3142 Background: Protein tyrosine kinase 7 (PTK7) is a highly conserved receptor tyrosine kinase involved in the Wnt signaling pathway and is overexpressed in multiple cancer types, including non-small cell lung cancer (NSCLC). Cofetuzumab pelidotin (ABBV-647) is an anti-PTK7 antibody-drug conjugate comprising the hu6MO24 monoclonal antibody, a cleavable cysteine-reactive linker, and Aur0101 (an auristatin microtubule inhibitor). It has shown promising preclinical anti-tumor effects (Damelin et al. Sci Transl Med 2017;9[372]:eaag2611) and clinical activity with a manageable safety profile in a Phase 1 study in patients with advanced solid tumors, with promising anti-tumor activity noted in NSCLC (Sachdev et al. DOI: 10.1200/JCO.2018.36.15_suppl.5565). Methods: This open-label, single-arm, multicenter Phase 1b study (NCT04189614) will assess the anti-tumor activity and safety of cofetuzumab pelidotin in approximately 40 patients with PTK7-expressing, recurrent NSCLC. The primary objective is to assess the objective response rate of cofetuzumab pelidotin according to Response Evaluation Criteria in Solid Tumors version 1.1. Secondary objectives include the duration of response, progression-free survival, overall survival, and safety and tolerability. Pharmacokinetic and biomarker samples will also be collected throughout for analysis. Patients must be aged ≥18 years with an Eastern Cooperative Oncology Group performance status of 0–1 and have recurrent histologically confirmed NSCLC with PTK7-expressing tumor (using a validated immunohistochemistry assay). Patients must have progressed after treatment with a platinum-based chemotherapy doublet and an immune checkpoint inhibitor (for tumors without targetable genetic alterations), or a platinum-based chemotherapy doublet and targeted agent(s) (for tumors with targetable genetic alterations). Patients must also have received ≤2 prior lines of systemic therapy (≤3 prior lines for tumors treated with targeted agent[s] for genetic alterations), including no more than 1 line of systemic chemotherapy. Cofetuzumab pelidotin (2.8 mg/kg) is administered intravenously every 3 weeks until the patient experiences disease progression, intolerable toxicity, or other study treatment discontinuation criteria are met. The study commenced on February 13, 2020 and enrollment is ongoing. Clinical trial information: NCT04189614.

9044 Background: Programmed death ligand-1 (PD-L1) expression is not a completely reliable predictive marker of the efficacy of anti-programmed cell death protein-1 (PD-1)/anti-PD-L1 therapy in advanced non-small cell lung cancer patients (NSCLC). Immune-related tumor microenvironment (TME) is classified into four different types based on the status of tumor-infiltrating lymphocytes (TILs) and PD-L1 expression. Methods: We retrospectively reviewed advanced NSCLC patients treated with anti-PD-1/anti-PD-L1 therapy between 2015 and 2019, and investigated the association between the efficacy of anti-PD-1/anti-PD-L1 therapy, the types of TME based on PD-L1 (clone: 22C3) expression, and the density of CD8-positive TILs by immunohistochemistry (/mm2), and mutational profiles assessed by next-generation sequencing. Results: Overall, 228 patients without driver mutation ( EGFR, ALK, ROS1, and RET) were included in the analysis. The patients were classified into four following groups: Type I: PD-L1High (tumor proportion score [TPS]≥50%)/TILHigh (≥85/mm2; n = 73), Type II: PD-L1Low (TPS < 50%)/TILLow ( < 85/mm2; n = 70), Type III: PD-L1High/TILLow (n = 37), and Type IV: PD-L1Low/TILHigh (n = 48). The progression-free survival (PFS) and objective response rate (ORR) of anti-PD-1/anti-PD-L1 therapy clearly differed according to the different tumor microenvironment (TME) types (ORR and median PFS; Type I: 64%, 14.5 months, Type II: 12%, 2.1 months, Type III: 24%, 3.6 months, Type IV: 41%, 10.8 months). In patients with PD-L1High tumors, Type I tumors had significantly better ORR and PFS than Type III(ORR: p < 0.001, and PFS: p < 0.001) tumors. Regarding the association between mutational profiles, histology and the TME types, the presence of TP53 mutation and KRAS mutation significantly related to TILHigh (Type I and IV) and PD-L1High tumors (Type I and III), respectively. Pleomorphic and NSCLC- not otherwise specified histology were associated with Type I tumors, while LCNEC was associated with PD-L1 low tumors (Type II and IV). Conclusions: Various factors (mutational profile and histology) are related to TME classification based on the status of TILs and PD-L1 expression. Differential types of TME, including PD-L1 expression and TILs status, can accurately predict the efficacy of anti-PD-1/anti-PD-L1 therapy.[Table: see text]

e21719 Background: Chemoradiotherapy (CRT) followed by durvalumab as maintenance therapy prolonged progression-free survival (PFS) and overall survival (OS) in unresectable locally advanced NSCLC (LA-NSCLC). Additionally, the history of radiotherapy and CRT has been reported to increase the efficacy of PD-1 blockade in advanced NSCLC patients. We evaluated the efficacy of anti-PD-(L)1 antibody therapy after CRT failure, and how CRT changes the status of PD-L1 expression on tumors and tumor infiltrated lymphocytes (TILs) in tumor microenvironment (TME) in unresectable LA-NSCLC patients. Methods: We retrospectively reviewed unresectable LA-NSCLC patients treated with CRT between December 2007 and December 2018, and evaluated the efficacy of PD-1 blockade after CRT failure. We also analyzed PD-L1 (clone: 22C3) expression on tumor cells, and CD8 positive TILs using the paired specimens that had been obtained pre-CRT and post-CRT failure. Results: We identified 422 patients who received CRT. Median follow-up was 36.1 months (range 2.7–138.1 months). Among these patients, sixty-five patients who had progressed post-CRT received anti-PD- (L)1 therapy (PD-1 therapy: 61 patients, PD-L1 therapy: 4 patients). Response rate (RR) and PFS of anti-PD-(L)1 therapy were 48% (95% CI, 35–60) and 8.7 (95% CI, 4.5–13.0) months. The RR and PFS did not differ according to PD-L1 expression levels (Table). Of the 18 patients, 9, 7, and 2 showed upregulation in PD-L1 expression or down- or no change, respectively, post-CRT. In contrast, the density of CD8 positive TILs in TME increased by CRT treatment ([pre-CRT]: median, 110 ± 239 /mm2 vs. [post-CRT]: median, 470 ± 533 /mm2, p = 0.025). Conclusions: The clinical outcome of anti-PD-(L)1 therapy after CRT failure in LA-NSCLC patients could be better than advanced NSCLC patients, but did not differ according to PD-L1 expression levels. The efficacy of PD-(L)1 therapy enhanced by CRT treatment could be due to the infiltration of CD8 T-cells into TME. [Table: see text]

44 Background: Cancer cachexia is known to a multifactorial catabolic syndrome and is observed 15-40% in general cancer population with worse prognosis. A recent study suggested a shorter survival period in non-small cell lung cancer (NSCLC) patients with a higher pembrolizumab clearance associated with cachexia. We here conducted analyses for the clinical impact of cachexia in advanced NSCLC treated with pembrolizumab. Methods: We evaluated consecutive advanced NSCLC patients who received pembrolizumab between March 2017 and December 2018 at the National Cancer Center Hospital. Cachexia was defined as a body weight loss >5% over the past 6 months or >2% in patients with a BMI

68 Background: Previous retrospective analyses have revealed higher response rates and a trend towards longer progression-free survival (PFS) and overall survival (OS) in response to immune checkpoint inhibitors (ICIs) in overweight patients. There are few reports concerning the association between the overweight state and the efficacy of ICIs specifically in non-small cell lung cancer (NSCLC) patients. We investigated the association between the body mass index (BMI) and the efficacy of ICIs in patients with NSCLC. Methods: Patients with advanced NSCLC who received ICI therapy (nivolumab or pembrolizumab) at the National Cancer Center Hospital from January 2016 to December 2018 were included in this retrospective cohort study. Based on their BMI, the patients were categorized into the overweight (A) group (BMI ≥25) and the non-overweight (B) group (BMI < 25). The PFS was compared between the two groups as the primary outcome. Results: Data of a total of 323 patients (median age, 63 years) were analyzed; 87 (26.9%)/43 (13.3%)/193 (59.8%) patients received pembrolizumab as 1st line therapy, pembrolizumab as 2nd line therapy, and nivolumab, respectively. Tumor proportion score was ≥50% in 58.4% (139/238) patients. The ECOG-PS was ≥2 in 37 patients (11.5%). The median body weight was 58.1, and the median BMI was 21.4. Of the 323 patients, 46 (14.2%) were categorized into group A (overweight) and 277 (85.8%) into group B (non-overweight). The median PFS and OS in two groups were as follows: A, 6.9 m/B, 5.6 m (HR 0.84, 95% CI [0.57-1.25], p = 0.38), and A, 22.3 m/B, 15.4 m (HR 0.90, 95% CI [0.56-1.43], p = 0.64), respectively. In accordance with the ICI regimen that the patients received, the PFS was A, 7.9 m/B, 7.8 m (HR 0.86, 95%CI [0.37-2.04], p = 0.74) in the patients who received pembrolizumab as 1st line therapy, A, 7.5 m/B, 5.3 m (HR 0.60, 95% CI [0.18-2.00], p = 0.40) in the patients who received pembrolizumab as 2nd line therapy, and A, 6.5 m/B, 4.4 m (HR 0.84, 95% CI [0.52-1.36], p = 0.48) in the patients who received nivolumab. Conclusions: Overweight NSCLC patients treated with ICIs showed a trend (non-statistically significant) towards a longer PFS as compared to non-overweight patients.

100 Background: Immunosenescence, an age-related impairment of the immune system, dampens acquired immunity and promotes inflammation. Therefore, it could also influence the degree of effectiveness of immune checkpoint inhibitors. Methods: We retrospectively reviewed the data of 52 NSCLC patients aged ≧75 years old treated with nivolumab or pembrolizumab from December 2015 to April 2019. Immunosenescence was assessed by the modified Glasgow Prognostic Scale (mGPS), Neutrophil-to-lymphocyte ratio (NLR), and Charlson Comorbidity Index (CCI), which are related to nutrition, inflammation and comorbidity, respectively. The Cox proportional hazard model and Kaplan-Meier curves were used to identify factors associated with survival. Results: The median follow-up duration was 19.5M (IQR:1-41). The patient characteristics were as follows: median age 79.7 years; male/female ratio, 41/11; PS0/1/2/3, 10/32/8/2; adeno/squamous histology/others, 33/15/4;stage III/IV/recurrence, 2/26/24; PD-L1 (22C3) (%)unknown/0/1-49/50-100, 18/2/8/24; driver mutation status positive/negative, 11/41; nivolumab/pembrolizumab, 29/23; treatment line 1/2/3/4 or more, 28/15/3/6; median number of treatment cycles 7.0 (1-53). The overall response rate (ORR) and disease control rate (DCR) were29.1% and 56.2%, respectively. The median progression-free survival (PFS) was 4.2 months (95% CI 1.8-7.5). The mGPS was significantly associated with the DCR (High/Low = 37.5/68.8%, p = 0.02) and the PFS (score 0-1/2 = 4.1/0.6 months) (HR: 0.37, 95% CI 0.18-0.74, p 3.5)/Low (≦3.5) = 3.8/5.6 months, p = 0.89). Multivariate regression analysis identified the mGPS as a significant predictor of the PFS(HR: 0.40, p = 0.008). Conclusions: A high mGPS score was significantly associated with a lower DCR and shorter PFS in elderly NSCLC patients treated with anti-PD-1 antibody.

8567Background: Prognostic factors and efficacy of chemotherapy remain unclear in patients with advanced thymic carcinoma. Methods: We performed a multi-institutional, retrospective study named NEJ...

e18556 Background: Thymic epithelial tumor, comprising of thymoma and thymic carcinoma, is a rare cancer, therefore, optimal chemotherapy for advanced thymic epithelial tumor is still controversial...