No abstract

No abstract

In recent years, immunotherapy-based regimens have been included into the treatment's algorithm of several cancer types. Programmed death-1 (PD-1) and cytotoxic T lymphocyte antigen-4 (CTLA-4) interact with their ligands found on the surface of antigen presenting cells (APC) or tumor cells (PD-L1/2 and CD80/86). Through these interactions, stimulatory or inhibitory signals are established. Immune checkpoint inhibitors (ICIs), block these interactions, and when administered not only as monotherapy but also as part of combination regimens, have shown to improve survival results in multiple advanced cancers leading to an increasing number of patients treated with ICI and, as a consequence, a rise in the number of patients developing immune-related adverse events (irAEs). Presence of irAEs has been associated with greater benefit from treatment, especially when blocking PD-L1. Recent data suggests that treatment benefit persists after discontinuation of ICIs due to a treatment related adverse event, regardless of the grade. Patients experiencing grade 3-4 irAEs are at risk of toxicity recurrence after reintroducing immunotherapy and therefore, the decision to resume the treatment is challenging. In these cases, a multidisciplinary approach is always needed and several factors should be considered. Management of severe toxicities may require systemic corticosteroids which can impact on T-cell function. Due to their immunosuppressive properties, it is necessary to deeper determine how corticosteroids influence responses. In terms of overall survival (OS), the use of steroids as therapy for irAEs seems not to reduce OS and several studies have reported durable responses in patients experiencing autoimmune toxicities treated with corticosteroids.

Immune checkpoint inhibitors (ICIs) have transformed non-small cell lung cancer (NSCLC) treatment. Unfortunately, only some patients benefit from these therapies. Thus, certain clinicopathological characteristics of the patients have been proposed as biomarkers of ICIs response. We assembled a retrospective cohort of 262 NSCLC patients treated with ICIs, compiled relevant clinicopathological characteristics, and studied their associations with treatment outcome using Cox proportional-hazards survival models. Additionally, we investigated the interrelations between clinicopathological features and devised a method to create a compendium associated with ICIs response by selecting those that provide non-redundant information. In multivariate analyses, ECOG performance status (hazard ratio (HR) 1.37 (95% CI 1.11 to 1.68), p < 0.005), LDH (HR 1.24 (95% CI 1.03 to 1.48), p = 0.02)) and PD-L1 negativity were associated with decreased PFS (HR 1.92 (95% CI 1.03 to 3.58), p < 0.04), whereas presentation of immune-related adverse events (irAEs) (HR 0.35 (95% CI 0.22 to 0.55, p < 0.005) or females (HR 0.52 (95% CI 0.33 to 0.80, p < 0.005) had longer progression-free survival. Additionally, numerous clinicopathological indicators were found to be interrelated. Thus, we searched for features that provide non-redundant information, and found the combination of LDH levels, irAEs, and gender to have a better association with ICIs treatment response (cross-validated c-index = 0.66). We concluded that several clinicopathological features showed prognostic value in our real-world cohort. However, some are interrelated, and compendiums of features should therefore consider these interactions. Joint assessment of LDH, irAEs, and gender may be a good prognostic compendium.

6081 Background: Treatment selection based on actionable alterations (AAs) is an appealing strategy for pts with R/M SGT. The GEMS-001 study (NCT02069730) at Princess Margaret Cancer Centre (PM) and the Vall D´Hebron Institute of Oncology (VHIO) pre-screening program facilitate the identification of AAs for R/M SGT pts and treatment selection. Methods: We analyzed R/M SGT treated at PM and VHIO from 2015 to 2020. Clinicopathological features, molecular alterations and treatment modalities were correlated with outcomes. The primary endpoint was overall response rate (ORR) by RECIST 1.1. Clinical benefit rate (CBR) was defined by pts with partial response or stable disease ≥4 months. Clinical actionability of multigene panel testing (NGS) and immunohistochemistry (IHC) were assessed as per institutional molecular tumor boards or investigators. Pts were opportunistically matched to available therapies from each center. Results: In total 206 pts were enrolled. On IHC, HER2 overexpression was present in 9%, Androgen Receptor (AR) 33%, Estrogen/Progesterone Receptor (ER/PR) 11% and ALK overexpression 0%. On NGS, PIK3CA mutation (mut) was in 9%, NTRK fusion 6%, NOTCH1-3 mut 5%, HRAS mut 6%, ERBB2/3 alterations (alt) 4% and FGFR1-4 alt 3%. Up to 92 pts (45%) displayed at least 1 AA and 36 pts (18%) had ≥2 AAs. A total of 60 pts (29%) were matched to AAs. Of those matched, median age was 60 years (range 33-84), M:F 21:39, 95% ECOG≤1 with a median number of prior treatment lines 0 (range 0-3), and their AAs included 26 AR, 9 HER2 or ERBB2 overexpression, 9 PIK3CA mut, 3 NTRK fusion, 3 FGFR1-3 alt and 10 other AAs (2 ER/PR overexpression, 2 EGFR mut, 1 c-kit mut, 1 BAP1 mut, 1 Non-V600 BRAF mut, 1 CDKN2A mut, 1 CHEK2 mut and 1 PTCH1 mut). Overall, ORR was 27% for the matched population. See table for outcomes. Conclusions: In our cohort, almost one third of the population received therapies matched to AAs. Our results suggest that targeted therapies have promising activity in pts with R/M SGT supporting comprehensive molecular and IHC profiling in treatment determination.[Table: see text]

CheckMate 743 trial demonstrated survival benefit of immunotherapy in first line in MPM with some differences in the efficacy of chemotherapy according to histology. The objective of this study is to characterize the impact of chemotherapy according to histology in patients diagnosed with MPM at our institution. Clinical records of all MPM patients diagnosed at Vall d’Hebron University Hospital between November 2002 and April 2020 were reviewed. Associations between clinical variables and outcomes were assessed with Cox regression models. Survival data were calculated by the Kaplan–Meier method. 189 patients were included with 76% of tumors classified as epithelioid subtype. First line chemotherapy was offered to 85% of patients. Median survival in overall population was 21.3 months (95% CI 17.2–24.3). We found that patients with epithelioid tumors had better overall survival (OS) and progression free survival (PFS). Median OS of epithelioid patients treated with first line chemotherapy was 26.7 months versus 15.0 months in non-epithelioid patients (HR 2.25 CI 95% 1.4–3.4; p < 0.001). Median PFS for patients with epithelioid tumors treated with chemotherapy was 4.8 months versus 3.6 months in non-epithelioid (HR 1.5 CI 95% 1.0–2.3; p = 0.03). The improvement of outcomes in patients with epithelioid histology was detected in patients treated with cisplatin or carboplatin. Histology was not a predictive factor for the platinum agent sensitivity (p of interaction PFS = 0.09, p of interaction OS = 0.65). In our series, patients with non-epithelioid tumors presented worse prognosis. Although epithelioid tumors exposed to cisplatin had higher PFS, histology was not a clear predictor of chemotherapy efficacy.

9071 Background: ICIs have significantly changed the therapeutic landscape of advanced NSCLC. As such, characterizing predictive markers of long-term clinical benefit is a critical objective. TMB quantification using targeted gene panels associates with long-term response to ICIs in NSCLC patients (Rizvi H ASCO 18). Although TMB quantified by targeted NGS correlates with that of WES, caution may be needed when using smaller panels. Methods: Here we analyzed WES of tumors and matched normal tissue from 67 NSCLC patients including 42 treated with ICIs. We correlated TMB with clinico-pathological features and outcomes. TMB was categorized as high vs. low according to the upper quartile of cohort distribution. Results: The median TMB was 2.68 non-synonymous variants (nSNVs)/Mb, ranging from 0 to 15.6 nSNVs/Mb, with upper quartile at 5.42 nSNVs/Mb. TMB was higher for smoker/current smoker (median 3.51) compared to never smokers (median 0.94, p = 0.0048) but no differences were seen in elderly ( > 70 years) vs. young patients or across histologies (squamous, adeno and other) and stages at diagnosis. In patients treated with ICIs, median TMB was 5.44 for those achieving complete response, 3.87 for patients with partial response and 2.42 for patients with progressive disease (PD) (p = 0.04). Moreover, improved clinical outcomes were associated with higher TMB (Table). In patients treated with ICIs, TMB as continuous variable had an impact on progression free survival (PFS) (p = 0.03). Median PFS was 22.3 months (mo) (14-not reached) for those with high TMB and 6.4 mo (3-16) for those with low TMB (HR 0.34, 0.13-0.9, p = 0.03). Median overall survival was not reached for those patients with high TMB and 32 mo (22-43) for those with low TMB (HR 0.29, 0.1-0.86, p = 0.02). Conclusions: High TMB correlates with long-term ICI benefit in NSCLC patients. Mutations in individual genes potentially linked to long-term benefit or resistance to ICIs will be presented. [Table: see text]

No abstract