Background/Objectives: Transanal irrigation (TAI) has been recognized as a safe and effective treatment for neurological bowel dysfunction, chronic constipation or fecal incontinence and has also been proposed for patients with low anterior resection syndrome (LARS). The aim of the present systematic review was to evaluate the feasibility and effectiveness of TAI in patients with significant LARS symptoms. Methods: We performed a systematic review according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement and guidelines in addition to the Cochrane Handbook for Systematic Reviews of Interventions. The protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) (CRD42023436839). The risk of bias was assessed using a modified version of the Downs and Black checklist. The main outcome was improvement in low anterior resection syndrome after TAI assessed by change in LARS score. Results: After an initial screening of 3703 studies, 9 were included and underwent qualitative synthesis (among them, 3 were randomized clinical trials). All studies recorded an improvement in LARS score following TAI procedure and almost all studies showed an improvement in other bowel function outcomes (Memorial Sloan Kettering Cancer Center Bowel Function Instrument (MSKCC BFI, ), Cleveland Clinic Incontinence Score (CCIS), visual analog scale (VAS), Cleveland Clinic Florida Fecal Incontinence Score (CCFFIS), fecal incontinence score (FI score), Obstructed Defecation Syndrome (ODS) score) and quality of life (QoL) scores. The discontinuation rate ranged from 0% to 41%. The rate of adverse events was high (from 0 to 93%); moreover, no uniformity was found in the various protocols used among the different studies. Conclusions: The results of this review show that TAI is effective in the treatment of LARS, improving the LARS score, the other bowel function outcomes and the QoL scores. The absence of a treatment protocol validated by the scientific community is reflected in the high disparity in terms of adverse events and discontinuation of therapy, in addition to representing an intrinsic limitation to the study itself.

The aim of this study was to evaluate the association of adipose tissue characteristics with survival in rectal cancer patients. All consecutive patients, diagnosed with stage II–IV rectal cancer between 2010–2016 using baseline unenhanced Computed Tomography (CT), were included. Baseline total, subcutaneous and visceral adipose tissue areas (TAT, SAT, VAT) and densities (TATd, SATd, VATd) at third lumbar vertebra (L3) were retrospectively measured. The association of these tissues with cancer-specific and progression-free survival (CCS, PFS) was assessed by using competitive risk models adjusted by age, sex and stage. Among the 274 included patients (median age 70 years, 41.2% females), the protective effect of increasing adipose tissue area on survival could be due to random fluctuations (e.g., sub-distribution hazard ratio—SHR for one cm2 increase in SAT = 0.997; 95%confidence interval—CI = 0.994–1.000; p = 0.057, for CSS), while increasing density was associated with poorer survival (e.g., SHR for one Hounsfield Unit—HU increase in SATd = 1.03, 95% CI = 1.01–1.05, p = 0.002, for CSS). In models considering each adipose tissue area and respective density, the association with CSS tended to disappear for areas, while it did not change for TATd and SATd. No association was found with PFS. In conclusion, adipose tissue density influenced survival in rectal cancer patients, raising awareness on a routinely measurable variable that requires more research efforts.

e13064 Background: The introduction in the clinical practice of CDK4 / 6 inhibitors, in combination with endocrine therapy, has been a breakthrough in the therapeutic scenario of hormone receptor+ve/Human Epidermal growth Factor 2–ve (HR+/HER2–) MBC pts. However, despite the outstanding results in terms of OS and quality of life achieved up to now, it is important to consider that the highly selected population of clinical trials doesn’t always accurately predict the clinical impact of therapies in the general population, hence having real world data is an essential tool to guide physicians in daily clinical practice. Methods: In this observational, multicentre, prospective study, we evaluated the safety profile and the incidence of dose reductions, temporary interruptions and treatment discontinuation due to adverse events, in a real–world cohort of Italian pre or postmenopausal HR+/HER2– MBC pts who received a first line combination of ribociclib and letrozole. Results: A total of 78 MBC patients were included in the analysis. The median duration of treatment was 17 (1–61) months. At the time of ribociclib initiation, 13 (16.7%) pts already presented an altered transaminase value whereas liver metastases were present in 17 pts (21.8%). Ten (12.8%) pts experienced hepatic toxicity while on treatment. The median number of cycles for liver toxicity was 3 (1–18). The most common hepatic event was transaminase level increase in 7 (8.8%) pts. Such event occurred more frequently in the group of pts with normal baseline liver function. Hepatic toxicity was grade 1 in 3 (3.8%) pts, grade 2 in 3 (3.8%) pts, grade 3 in 3 (3.8%) pts and grade 4 in 1 (1.3%) pt. Dose reduction due to hepatic toxicity occurred in 3 (3.8%) pts, in particular one of them needed dose reduction from 600 mg to 400 mg, one from 600 mg to 200 mg whereas treatment discontinuation was applied in the last patient. Median time of treatment interruption due to hepatotoxicity was 29 (14–45) days and 3 (3.8 %) pts discontinued ribociclib due to adverse events. No deaths due to liver failure have been reported. Conclusions: Hermione–8 trial support a manageable safety profile of the combination ribociclib and letrozole in MBC patients treated in a real–world clinical setting, confirming toxicity profile in line with the results presented in the MONALEESA clinical trials. Dose reductions, temporary interruptions and discontinuation of treatment due to adverse events have been limited and seems comparable to the results from MONALEESA trials. [Table: see text]

The study aimed to explore the impact of low skeletal muscle mass and quality on survival outcomes and treatment tolerance in patients undergoing radical chemo-radiation therapy for head and neck cancer (HNC). This is significant given the growing interest in sarcopenia as a possible negative predictive/prognostic factor of disease progression and survival. From 2010 to 2017, 225 patients were included in the study. Pre-treatment computed tomography (CT) scans of HNC patients undergoing (chemo)radiation therapy were retrospectively reviewed. The skeletal muscle area, normalized for height to obtain the skeletal muscle index (SMI), the skeletal muscle density (SMD) and the intramuscular adipose tissue area (IMAT) were measured at the level of the L3 vertebra. Low SMD and low SMI were defined according to previously reported thresholds, while high IMAT was defined using population-specific cut-point analysis. SMI, SMD, and IMAT were also measured at the proximal thigh (PT) level and tested as continuous variables. Clinical morpho-functional parameters, baseline nutritional markers with a known or suspected impact on HNC treatment, clinical outcomes and sarcopenia were also collected. In multivariate analyses, adjusted by age, sex, stage, diabetes, body mass index (BMI), and weight loss, L3-SMI was not significantly associated with survival, while poor muscle quality was negatively associated with overall survival (OS) (HR = 1.88, 95% CI = 1.09–3.23, p = 0.022 and HR = 2.04, 95% CI = 1.27–3.27, p = 0.003, for low L3-SMD and high L3-IMAT, respectively), progression-free survival (PFS) (HR = 2.26, 95% CI = 1.39–3.66, p = 0.001 and HR = 1.97, 95% CI = 1.30–2.97, p = 0.001, for low L3-SMD and high L3-IMAT, respectively) and cancer-specific survival (CSS) (HR = 2.40, 95% CI = 1.28–4.51, p = 0.006 and HR = 1.81, 95% CI = 1.04–3.13, p = 0.034, for low L3-SMD and high L3-IMAT, respectively). Indices at the PT level, tested as continuous variables, showed that increasing PT-SMI and PT-SMD were significant protective factors for all survival outcomes (for OS: HR for one cm2/m2 increase in PT-SMI 0.96; 95% CI = 0.94–0.98; p = 0.001 and HR for one HU increase in PT-SMD 0.90; 95% CI = 0.85–0.94; p < 0.001, respectively). PT-IMAT was a significant risk factor only in the case of CSS (HR for one cm2 increase 1.02; 95% CI = 1.00–1.03; p = 0.046). In conclusion, pre-treatment low muscle quality is a strong prognostic indicator of death risk in patients affected by HNC and undergoing (chemo)radiotherapy with curative intent.

TPS8591 Background: Surgery for malignant pleural mesothelioma (MPM) is indicated mainly in multimodal approaches and in clinical trial settings. Different studies have shown significantly lower complication rates, lower peri-operative morbidity and mortality with pleural/decortication (P/D), with similar overall survival rates. The biology of mesothelioma shows significant heterogeneity. Surgical tumor reduction may create a host environment more amenable to immunotherapy by reducing the ratio of tumor cells versus antitumor effector T lymphocytes, reducing the quantities of intratumor and/or systemic immunosuppressive cells, and ablating tumor-derived paracrine factors that promote local recruitment of immunosuppressive cells. Encouraging clinical data emerging in the field of tumor immunotherapy have demonstrated that therapies focused on enhancing T-cell responses against cancer can result in a significant survival benefit. The AtezoMeso Study evaluates the introduction of atezolizumab-ATEZO in MPM, patients (pts) after P/D and platinum/pemetrexed perioperative therapy. Methods: This is a double-blind, placebo controlled, phase III trial, in 20 Italian centers. Main inclusion criteria are P/D without macroscopic residual and ECOG-PS 0-1. Pts who underwent to P/D without macroscopic residual disease and have received at least 4 cycles of perioperative therapy with cisplatin/carboplatin and pemetrexed, will be randomized (2:1) to receive ATEZO or placebo. Therapy will be administered at dose of 1,200 mg iv, every 21 days, for 12 months or until recurrence, unacceptable toxicity or patient/physician decision. Randomization will be done through a centralized system, using histology (epithelioid vs nonepithelioid) and stage (I vs > I) as stratification factors. Pts will be radiologically evaluated after surgical before starting therapy and then every 12 weeks for 24 months or until recurrence. Quality of life (QoL) will be evaluated with the EQ-5D questionnaire administered at baseline and every 12 weeks. Tissue tumor samples will be centrally analyzed to determinate the genomic profile using FoundationOne CDx platform. The primary endpoint is the evaluation of the atezolizumab efficacy in terms of disease free survival (DFS). Secondary endpoints include the safety and efficacy in terms of overall survivall and QoL. Assuming a median DFS equal to 9 months in placebo arm, an accrual time of 24 months and a follow-up time of 24 months, a sample size of 162 pts will allow to detect true hazard ratios of 0.62 with power 0.8, at a confidence level of 95%. At February 14th, 2022, 3 pts have been enrolled. Clinical trial information: 2020-003762-39; GOIRC-02-2019.

Introduction: BRAF mutation involved 2–4% of lung adenocarcinoma. Differences in clinicopathologic features and patient outcome exist between V600E and non-V600E BRAF mutated NSCLC. Thus, we sought to assess the frequency and clinical relevance of BRAF mutations in a real-life population of advanced-NSCLC, investigating the potential prognostic significance of distinct genetic alterations. Materials and Methods: The present multicenter Italian retrospective study involved advanced BRAF mutant NSCLC. Complete clinicopathologic data were evaluated for BRAF V600E and non-V600E patients. Results: A total of 44 BRAFmut NSCLC patients were included (V600E, n = 23; non-V600E, n = 21). No significant differences in survival outcome and treatment response were documented, according to V600E vs. non-V600E mutations, although a trend towards prolonged PFS was observed in the V600E subgroup (median PFS = 11.3 vs. 6.0 months in non-V600E). In the overall population, ECOG PS and age significantly impacted on OS, while bone lesions were associated with shorter PFS. Compared to immunotherapy, first-line chemotherapy was associated with longer OS in the overall population, and especially in the BRAF V600E subtype. Conclusions: Here, we report on real-life data from a retrospective cohort of advanced-NSCLC harboring BRAF alterations. Our study offers relevant clues on survival outcome, therapeutic response, and clinicopathologic correlations of BRAF-mutant NSCLC.

Introduction Targeting Kirsten Rat Sarcoma (KRAS) has been deemed impossible for long time, but new drugs have recently demonstrated promising results. Evidence on the outcome of KRAS-mutant advanced-NSCLC treated with new standard regimens are still scarce. Thus, we aimed at assessing the incidence and clinical impact of KRAS mutations in a real-life population of advanced-NSCLC, exploring the prognostic significance of distinct alterations. Materials and Methods The present multicenter retrospective study, conducted by 5 Italian Centers from January 2018 to February 2020, involved 297 advanced KRAS mutant NSCLC. Complete clinico-pathological data were evaluated. Results Out of 297 patients, 130 carried KRAS_G12C mutation, while 167 presented with mutations other than G12C. Within KRAS_non-G12C group, 73%, 16.8% and 8.9% harboured G12X, codon 13 and Q61H alterations, respectively. No significant differences in survival outcome and treatment response were documented according to KRAS_G12C versus non-G12C, nor KRAS_G12C versus G12X versus other mutations. On univariate analysis ECOG PS, number and sites of metastatic lesions and PD-L1 status significantly impacted on survival. A clear trend towards worse prognosis was apparent in chemotherapy-treated patients, while immunotherapy-based regimens were associated to prolonged survival. Investigating the outcome of PD-L1 ≥ 50% population, we did not detect any significant difference between KRAS_G12C and non-G12C subsets. Conclusion Here, we report on real-life data from a large retrospective cohort of advanced NSCLC harbouring KRAS alterations, with particular attention to G12C mutation. Our study offers useful clues on survival outcome, therapeutic response and clinico-pathological correlations in KRAS-mutant setting, especially in the upcoming era of KRAS G12C targeting therapy.

Background There is a preclinical rationale for inhibiting angiogenesis in mesothelioma. We aimed to assess the efficacy and safety of the anti-VEGFR-2 antibody ramucirumab combined with gemcitabine in patients with pretreated malignant pleural mesothelioma. Methods RAMES was a multicentre, randomised, double-blind, placebo-controlled, phase 2 trial done at 26 hospitals in Italy. Eligible patients were aged 18 years or older, had Eastern Cooperative Oncology Group performance status 0-2, and histologically proven malignant pleural mesothelioma progressing during or after first-line treatment with pemetrexed plus platinum. Patients were randomly assigned (1:1) to receive intravenous gemcitabine 1000 mg/m2 on days 1 and 8 every 3 weeks plus either intravenous placebo (gemcitabine plus placebo group) or ramucirumab 10 mg/kg (gemcitabine plus ramucirumab group) on day 1 every 3 weeks, until tumour progression or unacceptable toxicity. Central randomisation was done according to a minimisation algorithm method, associated with a random element using the following stratification factors: ECOG performance status, age, histology, and first-line time-to progression. The primary endpoint was overall survival, measured from the date of randomisation to the date of death from any cause. Efficacy analyses were assessed in all patients who had been correctly randomised and received their allocated treatment, and safety analyses were assessed in all patients who received at least one dose of their assigned treatment. This trial is registered with ClinicalTrials.gov, NCT03560973, and with EudraCT, 2016-001132-36. Findings Between Dec 22, 2016, and July 30, 2018, of 165 patients enrolled 161 were correctly assigned and received either gemcitabine plus placebo (n=81) or gemcitabine plus ramucirumab (n=80). At database lock (March 8, 2020), with a median follow-up of 21.9 months (IQR 17.7-28.5), overall survival was longer in the ramucirumab group (HR 0.71, 70% CI 0.59-0.85; p=0.028). Median overall survival was 13.8 months (70% CI 12.7-14.4) in the gemcitabine plus ramucirumab group and 7.5 months (6.9-8.9) in the gemcitabine plus placebo group. Grade 3-4 treatment related adverse events were reported in 35 (44%) of 80 patients in the gemcitabine plus ramucirumab group and 24 (30%) of 81 in the gemcitabine plus placebo group. The most common treatment-related grade 3-4 adverse events were neutropenia (16 [20%] for gemcitabine plus ramucirumab vs ten [12%] for gemcitabine plus placebo) and hypertension (five [6%] vs none). Treatment-related serious adverse events were reported in five (6%) in the gemcitabine plus ramucirumab group and in four (5%) patients in the gemcitabine plus placebo group; the most common was thromboembolism (three [4%] for gemcitabine plus ramucirumab vs two [2%] for gemcitabine plus placebo). There were no treatment-related deaths. Interpretation Ramucirumab plus gemcitabine significantly improved overall survival after first-line standard chemotherapy, with a favourable safety profile. This combination could be a new option in this setting. Copyright (C) 2021 Elsevier Ltd. All rights reserved.

This systematic review with meta-analysis aimed to assess the effect of diffuse liver diseases (DLD) on the risk of synchronous (S-) or metachronous (M-) liver metastases (LMs) in patients with solid neoplasms. Relevant databases were searched for systematic reviews and cross-sectional or cohort studies published since 1990 comparing the risk of LMs in patients with and without DLD (steatosis, viral hepatitis, cirrhosis, fibrosis) in non-liver solid cancer patients. Outcomes were prevalence of S-LMs, cumulative risk of M-LMs and LM-free survival. Risk of bias (ROB) was assessed using the Newcastle-Ottawa Scale. We report the pooled relative risks (RR) for S-LMs and hazard ratios (HR) for M-LMs. Subgroup analyses included DLD, primary site and continent. Nineteen studies were included (n = 37,591 patients), the majority on colorectal cancer. ROB appraisal results were mixed. Patients with DLD had a lower risk of S-LMs (RR 0.50, 95% CI 0.34–0.76), with a higher effect for cirrhosis and a slightly higher risk of M-LMs (HR 1.11 95% CI, 1.03–1.19), despite a lower risk of M-LMs in patients with vs without viral hepatitis (HR 0.57, 95% CI 0.40–0.82). There may have been a publication bias in favor of studies reporting a lower risk for patients with DLD. DLD are protective against S-LMs and slightly protective against M-LMs for viral hepatitis only.

Background The liver is one of the most frequent sites of metastases in rectal cancer. This study aimed to evaluate how the development of synchronous or metachronous liver metastasis and overall survival are impacted by baseline liver steatosis and chemotherapy-induced liver damage in rectal cancer patients. Methods Patients diagnosed with stage II to IV rectal cancer between 2010 and 2016 in our province with suitable baseline CT scan were included. Data on cancer diagnosis, staging, therapy, outcomes and liver function were collected. CT scans were retrospectively reviewed to assess baseline steatosis (liver density < 48 HU and/or liver-to-spleen ratio < 1.1). Among patients without baseline steatosis and treated with neoadjuvant chemotherapy, chemotherapy-induced liver damage was defined as steatosis appearance, ≥ 10% liver volume increase, or significant increase in liver function tests. Results We included 283 stage II to IV rectal cancer patients with suitable CT scan (41% females; mean age 68 ± 14 years). Steatosis was present at baseline in 90 (31.8%) patients, synchronous liver metastasis in 42 (15%) patients and metachronous liver metastasis in 26 (11%); 152 (54%) deaths were registered. The prevalence of synchronous liver metastasis was higher in patients with steatosis (19% vs 13%), while the incidence of metachronous liver metastasis was similar. After correcting for age, sex, stage, and year of diagnosis, steatosis was not associated with metachronous liver metastasis nor with overall survival. In a small analysis of 63 patients without baseline steatosis and treated with neoadjuvant chemotherapy, chemotherapy-induced liver damage was associated with higher incidence of metachronous liver metastasis and worse survival, results which need to be confirmed by larger studies. Conclusions Our data suggest that rectal cancer patients with steatosis had a similar occurrence of metastases during follow-up, even if the burden of liver metastases at diagnosis was slightly higher, compatible with chance.

Epidermal growth factor receptor T790M detection using liquid biopsy was evaluated in a real-life setting in 120 advanced nonesmall-cell lung cancer patients whose disease had progressed during first- or secondgeneration tyrosine kinase inhibitors. The T790M detection rate was 25.8% using liquid biopsy and 49.2% after tissue rebiopsy. Liquid biopsies performed before disease progression according to Response Evaluation Criteria In Solid Tumors were all negative for T790M and T790M positivity was higher in cases of extrathoracic metastatic sites. Background: In advanced epidermal growth factor receptor (EGFR)-mutated nonesmall-cell lung cancer (NSCLC) patients whose disease has progressed during treatment with first- and second-generation tyrosine kinase inhibitors (TKIs), liquid biopsy (LB) is routinely used to evaluate the presence of EGFR T790M as an acquired resistance mechanism. The objective of this study was to assess a real-life picture of EGFR T790M detection in LB. Materials and Methods: Liquid biopsies performed between June 2016 and October 2018 for advanced EGFR-mutated NSCLC at disease progression during treatment with first- and second-generation TKIs were retrospectively evaluated in 5 Italian centers. Circulating tumor DNA was extracted from plasma and tested with different commercial kits. The detection rate in LBs and the patients' characteristics were correlated. Results: We enrolled 120 consecutive patients. The overall T790M detection rate observed using LB was 25.8%. Fifty-four of 89 (60.7%) patients with negative LB results underwent tissue rebiopsy, and 56% were positive for T790M. The overall rate of T790M positivity in the study cohort was 49.2%. LB performed before formal tumor progression according to Response Evaluation Criteria In Solid Tumors criteria was negative for T790M in all patients (n=21; P=.012). T790M positivity was statistically significantly higher in cases of disease progression at extrathoracic metastatic sites (P=.008) and, specifically, in the case of worsening bone disease (P=.003). Conclusion: Our study shows that the detection of T790M-positive patients whose disease progressed during treatment with first- and second-generation TKIs in real life was according to the literature. However, this result was obtained with a specific clinical course (repeat LBs and tissue rebiopsy), thus implying the necessity for multidisciplinary management. (C) 2020 Elsevier Inc. All rights reserved.

9004 Background. The RAMES Study (EudraCT Number 2016-001132-36) is a multicenter, double-blind, randomized phase II trial exploring the efficacy and the safety of the addition of ramucirumab to gemcitabine as the second-line treatment in MPM patients (pts) after platinum/pemetrexed regimens. Methods. The pts were assigned (1:1) to receive Gemcitabine 1000 mg/m 2 i v on days 1 and 8 every 21 days with Placebo (Arm A) or Ramucirumab 10 mg/kg i v on day 1, of a 21-day cycle (Arm B), until tolerability or progressive disease. Pts was stratified by ECOG/PS (0-1 vs 2), age (≤ 70 vs > 70 yrs), histology (epithelioid vs non-epithelioid) and time to progression (TTP) after first-line therapy. The primary endpoint was overall survival (OS). Assuming a proportion of OS equal to 40% at 1 year in arm A, a 12% absolute improvement in OS at 1 yrs was expected in Arm B (hazard ratio = 0.70).114 events (156 subjects) are required for a one-sided log-rank test with α = 0.15 to have 80% power. Results. From December 2016 to July 2018, 164 pts were randomized, 81 pts in Arm A and 80 Arm B; 3 pts were randomized but not treated. Characteristics of pts were: median age 69 yrs (44-81), males 119 (73.9%), females 42 (26.1%); ECOG/PS0 96 (59.6%) ECOG/PS1-2 65 (40.4%); histotype epithelioid 132 (81.9%), non-epithelioid 29 (18.1%); stage III 98 (60.7%), stage IV 60 (37.3%), 3 (2.0%) missing; asbestos exposure assessed 80 (49.7%). Median of courses was 3.50 in Arm A and 7.50 in Arm B. OS was significantly longer in Arm B with median 13.8 mths (70% CI 12.7-14.4) vs Arm A with 7.5 mths (70% CI 6.9-8.9), HR 0.71 (70% CI 0.59-0.85, p = 0.057). OS at 6 and 12 mths was in Arm A 63.9% and 33.9%, and in Arm B 74.7% and 56.5%, respectively. In Arm B OS was not correlated to TTP at first-line therapy (13.6 mths in TTP ≤6 mths and 13.9 mths in TTP > 6 mths) and histotypes (13.8 months in the epithelioid and 13.0 months in non-epithelioid). No significant differences in thromboembolism G3-4 events were observed in Arm A vs Arm B (p= 0.64). None hypertension G3-4 was reported in Arm A vs 5 pts (6.3%) in Arm B (p= 0.022). No significant differences in G3-4 haematological toxicities between the two arms were reported. Conclusion: In the RAMES Study the addition of Ramucirumab to Gemcitabine significantly improved OS regardless of age of pts, tumor histotype and TTP at the first-line treatment. Gemcitabine plus Ramucirumab can be considered a manageable regimen in second-line treatment of advanced MPM pts. Clinical trial information: NTC03560973.

Purpose: Malignant pleural mesothelioma (MPM) is an aggressive cancer. Data are not available in prospective trials on correlations between genetic alterations and outcomes of therapies. In this study, we assessed the genetic profile of MPM patients (pts) in tissue samples. Patients and Methods: From December 2016 to July 2018 (end of enrolment), 164 pts were enrolled. We evaluated by targeted sequencing the mutational profile of a panel of 34 genes: ACTB, ACTG1, ACTG2, ACTR1A, BAP1, CDH8, CDK4, CDKN2A, CDKN2B, COL3A1, COL5A2, CUL1, DHFR, GOT1, KDR, KIT, MXRA5, NF2, NFRKB, NKX6-2, NOD2, PCBD2, PDZK1IP1, PIK3CA, PIK3CB, PSMD13, RAPGEF6, RDX, SETDB1, TAOK1, TP53, TXNRD1, UQCRC1, XRCC6. Genetic profiling was correlated with clinical and pathological variables. Results: Overall, 110 pts (67%) from both treatment arms had samples available for molecular analysis. Median age was 63 years (45–81), 25.5% (n = 28) were females, and 74.5% (n = 82) were males. Tumor histotype was 81.8% (n = 90) epithelioid and 18.2% (n = 20) non-epithelioid; 28.5% of the tumors (n = 42) were stage IV, 71.5% (n = 68) were stage III. Targeted sequencing of tissue specimens identified 275 functional somatic mutations in the 34 genes analyzed. The number of mutated genes was positively associated with higher stage and metastatic disease (p = 0.025). RDX (42%), MXRA5 (23%), BAP1 (14%), and NF2 (11%) were the most frequently mutated genes. Mutations in RAPGEF6 (p = 0.03) and ACTG1 (p = 0.02) were associated with the non-epithelioid subtype, and mutations in BAP1 (p = 0.04) were related to progression-free survival (PFS) > 6 months. Conclusions: In the Ramucirumab Mesothelioma clinical trial (RAMES), mutation of the gene BAP1 is related to a prolonged PFS for patients treated with platinum/pemetrexed regimens (p = 0.04).

e14156 Background: Checkpoint inhibitors in cancer therapy have shown survival benefits in the treatment of several ATs. However, the risk of serious side effects seems greater in certain types of tumors presenting reduced compliance with therapy. The aim of this study is to evaluate the safety profile and clinical benefits of IT, as appraised using clinically related biomarkers of pts outside clinical trials. Methods: In this retrospective study, data were evaluated on pts in the Reggio Emilia Provincial Oncology Network who were treated for ATs using Nivolumab, Pembrolizumab and Ipilimumab monotherapy in clinical practice. The pts included in the study had received at least 1 dose of IT by June 2018. LDH and neutrophil–lymphocyte ratio (NLR) were collected at baseline, antibiotic therapy (ATB) was considered during IT, and immunorelated adverse events (irAEs) were assessed with CTCAE v.4.1. Results: A total of 169 pts were examined: 78 with NSCLC, 57 with melanoma, 29 with kidney cancer, 3 with head-neck cancer and 2 with colorectal cancer. 123 pts (73%) were treated with Nivolumab, 36 (21%) with Pembrolizumab and 10 (6%) with Ipilimumab. Overall, 62 pts (37%) experienced irAES of any degree: 17/29 (59%) pts with kidney cancer presented irAES, while only 35% (20/57) and 31% (24/78) of pts with melanoma and lung cancer did, respectively. A total of 27/62 (43%) pts delayed IT due to toxicity, but only 7/62 (11%) pts discontinued treatment due to irAEs. Statistically significant increases in PFS (11 vs. 5 months, p = 0.002) and OS (19.9 vs. 8.7 months, p = 0.002) were found between pts who experienced irAEs and those who did not. As previously demonstrated in other groups of pts, in the Cox regression performed for OS, a baseline value of LDH≥1.5xULN and median NLR ≥2.93 correlates with poor prognosis. Unlike what is already known, the use of ATB during IT (35% of pts) correlates with increased survival (HR 0.59; IC95% 0.37-0.94, p = 0.028). Conclusions: These data confirm the tolerability of IT in an unselected real-world population. In particular, the development of irAEs could be considered a positive predictor of outcome, but with a different incidence among MTS.

e15175 Background: Despite the combination of fluoropyrimidine chemotherapy and long course radiotherapy (50.4 Gy) followed by TME, LARC still results in 30% distant recurrence within 2-3 years after surgery. One of the most affected organs is the liver. We aimed to analyze if basal liver pathological alterations and drug-induced liver damages influence the liver recurrence. Methods: 50 patients (pts) with LARC treated with neoadjuvant chemoradiotherapy (NACRT) between June 2010 and June 2017 in a single center were included. Liver function tests (LFTs) including AST, ALT and GGT at baseline and after NACRT were evaluated. The first staging CT scan and, when available, post-NACRT CT scan were reviewed to assess steatosis (unenhanced liver density < 48 HU or liver-to-spleen ratio < 1.1), and liver volume. Associations of baseline LFTs and CT liver alterations and changes after NACRT with liver metastasis (LM) during follow-up were evaluated with Hazard Ratio (HR) and respective 95%CI calculated with Cox analysis. Results: The characteristics of pts were: 68% males, median age 60.4 ys (range 34-84), cTN stage II (32%) and stage III (64%). During a follow-up ranging from 1.5 to 7 ys, 17(34%) pts developed distant recurrence and 7(14%) LM. At baseline, 2 pts had GGT > 73 U/l and 11(22%) had steatosis. After NACRT, 4(9%) pts had AST > 40 U/I and/or ALT > 49 U/l with > 2-folds increase, 6 (40%) of the 15 pts with available unenhanced post-NACRT CT had steatosis development, or increase (≥10 HU density decrease), and 9 (27%) of 33 pts with available post-NACRT CT had a > 10% increase in liver volume. LM was associated with female sex (p = 0.016). Baseline steatosis, post-NACRT steatosis development or increase, > 10% liver volume and transaminase increase were associated with LM (HR = 3.36, 95%CI = 0.75-15.1; HR = 3.67, 95%CI = 0.33-40.7; HR = 2.91, 95%CI = 0.59-14.5; HR = 3.05, 95%CI = 0.33-27.72, respectively). All the reported associations almost disappeared, when considering all-sites of distant recurrence. Conclusions: Even if small numbers cannot exclude that associations are due to chance, baseline liver steatosis and post-NACRT liver damage may be involved in LM during follow-up of LARC.

Background Deleterious polymorphisms in the gene encoding DPD (DPYD) may result in severe reduction of DPD enzymatic activity that causes life-threatening toxicities when the standard dose of fluorouracil is used. The best panel of single-nucleotide polymorphism (SNPs) of DPYD is not well defined. Methods In 2011, we began screening DPYD*2A in patients candidate for fluoropyrimidine-based chemotherapy. We planned a case-control study with all cases of DPYD*2A wild type who developed toxicity ≥G3 and with a cohort of patients who did not present severe toxicities. Then, we tested the additional SNPs: c.2846A>T, c.1679T>G, c.2194G>A. Results From 2011 to 2016, we screened 1827 patients for DPD deficiency; of those, 31 subjects (1.7%) showed DPYD*2A SNP. We selected 146 subjects who developed severe toxicities (Cases) and 220 patients who experienced no or mild toxicities (Controls); 53 patients carried one of the additional SNPs: 35 subjects (66%) fell into the Cases and 18 (34%) into the Controls ( p < 0.0001). c.2194G>A was the most frequent SNP (12.5%) and showed a correlation with neutropenia. We confirmed that c.2846A>T and c.1679T>G were related to various toxicities. Conclusions The additional DPYD polymorphisms could enhance the prevention of fluoropyrimidine toxicity. c.2194G>A is the most frequent polymorphism and it was found to be associated with neutropenia.

8547 Background: Since MPM is an uncommon neoplasia, its rarity has limited available data on molecular drivers. Methods: RAMES study evaluated the second-line efficacy of gemcitabine/ramucirumab treatment vs. gemcitabine/placebo. From December 2016 to July 2018 (end of enrolment), 164 patients (pts) were admitted to this study, which involved the collection of tumor samples - with diagnosis - to evaluate 34 genes by NGS (ACTB, ACTG1, ACTG2, ACTR1A, BAP1,CDH8, CDK4, CDKN2A, CDKN2B, COL3A1, COL5A2, CUL1, DHFR, GOT1, KDR, KIT, MXRA5, NF2, NFRKB, NKX6,-2, NOD2, PCBD2, PDZK1IP1, PIK3CA, PIK3CB, PSMD13, RAPGEF6, RDX, SETDB1, TAOK1, TP53,TXNRD1, UQCRC1, XRCC6). We reported the results of the first 87 pts (54%): hystotype was epithelioid in 70 pts (80%), biphasic in 14 pts (16%) and sarcomatoid in 3 pts (4%). Median age was 63 years (range 45-81). 70 pts were male (80%) and 17 pts were female (20%). In the present analysis, we included 55 pts in stage III (63%), 26 pts in stage IV (30%) and 6 pts whose stage was unknown. Median first-line PFS platinum/pemetrexed therapy was for 5.75 months (I.C. 95% 4.75-6.76). PFS was ≤6 months for 40 pts (49%), and 6 months for 41 pts (51%). Results: 187 functional somatic mutations were identified. Genomic alterations/patient were 1 gene in 29 pts (33%), 3 genes in 18 pts (21%) and ≥5 genes in 2 pts (2%). The most frequent somatic mutations were RDX in 35 pts (40%), MXRA5 in 20 pts (23%), BAP1 in 13 pts (15%) and ACTG 1 in 9 pts (11%). When patients were collated by stage, the most frequent mutations were: MXRA5 in 16 pts in stage III (29%), BAP1 in 5 pts in stage IV (19%) and RDX in 16 pts in stage IV (62%). The percentage of somatic mutations in patients with PFS as first-line chemotherapy for ≤6 and >6 months was 2.2 and 1.6 (p=0.032), respectively. The most frequent mutations/patient for ≤6 and >6 months PFS were: RDX in 14 pts (35%) with PFS < 6, RDX in 19 pts (46%) with PFS >6 and MXRA5 in 11 pts (27%) with PFS >6. Conclusions: This preliminary data suggests a possible role that a genetic signature may play in distinguishing MPM with different clinical-pathological features. The results are expected to be clarified further in the second step of the study, which is ongoing. Clinical trial information: 2016-001132-36.

e16067 Background: HER2 is an important biomarker in advanced gastric cancer (GC). Adding trastuzumab to chemotherapy (CHT) improved survival in patients with metastatic GC with overexpression of HER2. In the first step of the study we intend to describe the main clinical characteristics and outcome of pts with HER2 positive GC. Methods: Consecutive cases of HER2 positive GC observed at the Reggio Emilia Clinical Cancer Centre were evaluated from 2009 to 2017. We collected their pathological and clinical data. IHC for HER2 overexpression and FISH for gene amplification were used. Results: We analysed the HER2 expression in 443 pts with GC, 135 (30%) pts resulted HER2 positive, 93 men (69%) and 42 women (31%), with a mean age of 67 years (range 37-86). There was a concordance of 87% in 94 pts evaluated with both IHC and FISH. We found IHC 3+ in 63 pts (47%), IHC 2+ and FISH+ in 51 pts (38%), IHC- and FISH+ in 10 pts (7%) and only FISH+ (IHC not evaluated) in 10 pts (7%). We excluded 10 pts who did not present a metastatic disease from analysis. The primary tumour site was gastric in 76 pts (62%), gastroesophageal junction in 45 pts (35%) and unknown in 4 pts (3%). Of 125 pts, 95 (76%) presented synchronous metastasis and 30 (24%) metacronous metastasis; 66% of pts have 1 site of metastasis and 34% have multiple site of metastasis. 99 pts (79%) were treated with a first line of CHT and 95 (96%) of them received platinum/fluopirymidines regimen and trastuzumab with an 8-cycle mean duration of treatment (range 1-50). 48 pts (38%) received a second line of CHT and 16 (12%) a third line. 21 pts were not treated: 14 pts died early due to poor clinical condition and we have no data for 7 pts. See the table for survival analysis. Conclusions: In our single centre series we observed an incidence of 30% of HER2 positive GC. Almost all pts (96%) received first line trastuzumab-based CHT regimen with a longer survival in a subgroup of subjects that completed more than 6 cycles of therapy. In these responsive pts the research of biomarkers predictive of response is ongoing. HER2 positive 125 pts Treated with Trastuzumab 95 pts Trastuzumab < 6 cycles 44 pts Trastuzumab ≥6 cycles 42 pts OS % % % % 1 yr 38.5 47.7 25 76.2 2 yrs 12 14.7 2.3 28.6 3 yrs 6.5 7.4 0 16.7

e20524 Background: over the past decade, personalized management based on the molecular features of tumours in patients with advanced non small-cell lung cancer (NSCLC) has entered routine clinical practice. The poor performance of many advanced NSCLC patients may limit invasive biopsies. The liquid biopsy is a diagnostic procedure performed on cancer-derived material obtained in blood samples. In this abstract, we will describe our experience with liquid biopsies. Methods: In the Reggio Emilia Clinical Cancer Centrefrom March 2016 to December 2016, 42 patients with advanced NSCLC were analyzed that had had or had already started first line therapy. The liquid biopsy was repeated at each imaging response evaluation by thoracic-abdominal compound tomography (CT) scan performed every 3 months. In the liquid biopsy, the mutational status of EGFR was analyzed with real time PCR (KIT cobas EGFR mutation test v2 CE-IVD Roche); in tissue, it was evaluated by pyrosequencing. Results: 21/42 liquid biopsies were EGFR-mutated (12/21 eson 19 and 9/21 eson 21). In 3/21 (14.3%) cases, the tissue biopsies showed wild type (WT) EGFR. 6 liquid biopsies were also performed at time 0 (diagnosis). All liquid biopsies of EGFR WT remained WT during treatment and imaging evaluation. The median number of liquid biopsy tests for patients was 2 (range 1-3). In 4/21 cases, T790M was performed: 3 cases in both liquid biopsies and tissue, and 1 case in tissue but not in liquid biopsy. TKi therapy was ineffective in this patient with T790M mutation detected in tissue, but not in liquid biopsy. In all patients, the disappearance of the T790M mutation during TKi therapy was related to disease progression. In 11 cases, modification of EGFR mutation status during treatment anticipated CT scan evidence of disease progression (median = three months). Conclusions: the liquid biopsy is an excellent resource. In our experience the liquid biopsy is the sensitive method of choice during treatment of advanced NSCLC patients. EGFR modification status during TKi therapy showed advanced disease progression.

2567 Background: Dihydropyrimidine dehydrogenase (DPD) is a key enzyme in the metabolism of fluorouracil. Deleterious polymorphisms in gene-encoding DPD ( DPYD) results in a DPD deficiency that causes life-threatening toxicities when the standard dose of fluorouracil is used. DPYD*2A (IVS14+1G > A) is the most common single-nucleotide polymorphism (SNP) associated with critical DPD deficiency. At present, most of the evidence supports screening for at least 3 SNPs (DPYD*2A, c.2846 A > T, c.1679T > G). The aim of this study is to confirm that the detection of additional polymorphisms of DPYD could enhance prevention of fluoropyrimidine toxicity. Methods: In 2011, we began to screen DPYD*2A in patients candidate for fluoropyrimidine based-chemotherapy. As the first step of the evaluation, we selected all cases of DPYD*2A wild type, from 2011 to 2012, who developed CTC-NCI-V.3 toxicity ≥ G3. In these patients, we researched the other 3 SNPs (c.2846 A > T, c.1679T > G, c.2194C > A). Mutational status was analyzed with real Time PCR. Results: From 2011 to 2016 we pre-emptively screened DPD deficiency in 1,863 patients and 32 subjects (1.6%), with results mutated for DPYD*2A. As the first step of the evaluation, 548 subjects were assessed from 2011 to 2012. We found 7 patients who were carriers of the DPYD*2A mutation (1.27%). Of the 541 wild type cases, 114 presented toxicities ≥ G3. In this subgroup, 22 patients (19%) proved to be mutated for the other SNPs of DPYP, as reported in the table below. Conclusions: Preliminary data show that in 22 (19%) of 114 patients who presented severe toxicity which was not correlated with DPYD*2A, we found other polymorphisms of gene encoding DPD. Out of the 3 SNPs evaluated, c.2194 C > A proved to be the most frequent, although it is the polymorphism that is least known and least studied. Such results suggest that the evaluation of additional polymorphisms could enhance the prevention of fluoropyrimidine toxicity. The results are expected to be clarified further in the second step, which is ongoing. [Table: see text]