Search

Background and aim: Sorafenib is a multikinase inhibitor approved for the treatment of advanced hepatocellular carcinoma (HCC). However, the predictive factors of outcome among treated patients have not been fully investigated. Aim of this study was to identify factors predictive for tumor control among patients with advanced HCC treated with sorafenib. Material and methods: All consecutive HCC patients treated with sorafenib at our Unit from October 2008 to October 2010 were evaluated. All patients received 800 mg/day of sorafenib until progression or unacceptable toxicities. Overall survival (OS) and time to progression (TTP) were estimated by using Kaplan-Meier curves. TTP was considered as a marker of tumor control. Univariate analysis included age, gender, etiology, AFP, Child-Pugh class, MELD, alanine aminotransferase (ALT), alkaline phosphatase, bilirubin, albumin, ascites, metastasis, portal vein thrombosis, oesophageal varices, effects related to VEGF inhibition: rash, arterial hypertension, bleeding, and hand-foot syndrome (HFS). To detect factors independently related to TTP, multivariate analysis by Cox’s model was done. Results: We treated 101 patients (median age 67 years; 81 males; 98 cirrhotics; Child A/B: 79/19; 55 macrovascular invasion; 39 extrahepatic spread). Median therapy duration was 4 (range 1-16) months. Median OS was 10 (95% CI 4.5-15.5) months. TTP was 9 (95%CI 3.7-14.3) months. Complete/partial response occurred in 21% of patients and stable disease in 19%. At univariate analysis, the variables significantly related to TTP were: female sex (14 months vs 6 months, p=.05), no portal thrombosis (13 months vs 6 months, p=.04), ALT =60UI (15 months vs 6 months, p=.02), no ascites (11 months vs 2 months, p=.003), and occurrence of HFS (16 months vs 5 months, p

e20558 Background: Recent literature suggests that ACT near the end of life is increasingly common. Although the reason is complex, we hypothesize that AC impair the formulation of prognosis. The PaP score was shown to predict 1-month survival for terminally ill pts. Its application to pts with AC is object of debate. We evaluated the frequency of ACT near the end of life and assessed the value of PaP score and of other clinical parameters in predicting survival in pts with AC. Methods: All pts admitted due to AC in a Department of Medical Oncology in a 9-month period were included. Pts and disease features (site of primary, stage of disease, disease control), previous and current ACT, cause of admission (diagnosis without a previous history of cancer; treatment-related toxicity; cancer- related symptoms) were assessed. The PaP score was calculated using physicians’ survival prediction, Performance Status (PS), anorexia and dyspnea, total WBC and lymphocyte percentage. Results: 208 pts were evaluated: local disease 24%, advanced 76%; ACT 60%, off-treatment 40%; disease controlled 9%, non controlled 91%; cause of admission: diagnosis 19%, toxicity 17%; symptoms 64%. PaP score: group A 78%, B 18%, C 4%. Overall, median survival was 19 weeks (95% I.C. 13,3 to 24,6) and 12-week survival was 59,6%. Of the pts admitted during ACT 11% died within 4 weeks and 32% within 12 weeks. Among the 39 pts died within 4 weeks, 33% were on ACT. Cause of admission, disease control, treatment, and PaP score predicted 12-week survival in the multivariate analysis. However, only PS < 30, and PaP group B and C identified pts with a 12-week survival lower than 30%. Conclusions: ACT in the last period of life is common. During AC only a minority of pts has such a dismal prognosis that treatment withdrawal is justified. The potential reversibility of AC renders prognostic measures inadequate to make treatment decisions. No significant financial relationships to disclose.

are presented in tables 1 and 2. Serious adverse events occurred in 12 patients of the budesonide group and 14 patients of the prednisone group. Conclusion: In this largest AIH-trial, ever conducted, 3mg tid oral budesonide is superior to oral prednisone each in combination with AZA in inducing biochemical remission in active autoimmune hepatitis with significantly less steroid specific side effects (SSSE).

10086 Background: Chemo-radiotherapy results in clinical cure for stage III/IV head and neck cancer in approximately 40% of cases, with significant treatment-associated morbidity and mortality. Molecular genetic factors predictive of treatment outcome would clearly be of value in selection of patients with highest probability of response. We have analysed the structure and epigenetic regulation of specific genes as possible predictors of outcome to chemo-radiotherapy. The genes analyzed were: p53 (single nucleotide polymorphism (SNP) and mutation), MDM2 (SNP), Chfr (methylation), CRABP1 (methylation). Methods: 84 patients with locally advanced head and neck cancer receiving cisplatin-based chemo-radiotherapy were studied. SNP genotypes were determined by direct sequencing of DNA from normal tissue. Acquired mutations in p53 and aberrant methylation in the CpG islands of specific genes were analyzed by direct sequencing and methylation-specific PCR. Results: There were 6 treatment-related deaths in 84 patients, all occurring in cases with germ-line haplotype p53 72 Arg/Arg, MDM2 309T/T. At the time of analysis, 39/84 (46%) patients had progressed or died, with median time to progression or death = 17.3 months. Complete response was more common in patients whose genetic/epigenetic profile was p53 72 Arg/Arg wild type, MDM2 309 G/G, Chfr methylated, CRABP1 methylated, compared to the profile p53 72 Arg/Arg mutant, MDM2 309 T/T, Chfr unmethylated, CRABP1 unmethylated (91% vs 46%, log rank p = 0.002). Progression-free survival was significantly longer for patients with the profile p53 72 Arg/Arg wild type, MDM2 309 G/G, Chfr methylated, CRABP1 methylated, compared to those with p53 72 Arg/Arg mutant, MDM2 309 T/T, Chfr unmethylated, CRABP1 unmethylated (% surviving progression-free at 2 years = 74% vs 36%, p = 0.001). Overall survival was significantly longer in patients with the profile p53 72 Arg/Arg wild type, MDM2 309 G/G, Chfr methylated, CRABP1 methylated compared to those with p53 72 Arg/Arg mutant, MDM2 309 T/T, Chfr unmethylated, CRABP1 unmethylated (% surviving at 2 years = 88% vs 40%, p = 0.0004). Conclusions: Genetic and epigenetic parameters influence the toxicity and clinical outcome of chemo-radiotherapy in head and neck cancer. No significant financial relationships to disclose.

Purpose: Metronomic capecitabine (MC) is a well-tolerated systemic treatment showing promising results in one retrospective study, as second-line therapy after sorafenib failure, in patients with hepatocellular carcinoma (HCC). Methods: 117 patients undergoing MC were compared to 112 patients, eligible for this treatment, but undergoing best supportive care (BSC) after sorafenib discontinuation for toxicity or HCC progression. The two groups were compared for demographic and clinical features. A multivariate regression analysis was conducted to detect independent prognostic factors. To balance confounding factors between the two groups, a propensity score model based on independent prognosticators (performance status, neoplastic thrombosis, causes of sorafenib discontinuation and pre-sorafenib treatment) was performed. Results: Patients undergoing MC showed better performance status, lower tumor burden, lower prevalence of portal vein thrombosis, and better cancer stage. Median (95% CI) post-sorafenib survival (PSS) was longer in MC than in BSC patients [9.5 (7.5–11.6) vs 5.0 (4.2–5.7)months (p < 0.001)]. Neoplastic thrombosis, cause of sorafenib discontinuation, pre-sorafenib treatment and MC were independent prognosticators. The benefit of capecitabine was confirmed in patients after matching with propensity score [PSS: 9.9 (6.8–12.9) vs. 5.8 (4.8–6.8)months, (p = 0.001)]. MC lowered the mortality risk by about 40%. MC achieved better results in patients who stopped sorafenib for adverse events than in those who progressed during it [PSS: 17.3 (10.5–24.1) vs. 7.8 (5.2–10.1) months, (p = 0.035)]. Treatment toxicity was low and easily manageable with dose modulation. Conclusions: MC may be an efficient and safe second-line systemic therapy for HCC patients who discontinued sorafenib for toxicity or tumor progression.