Background: There is limited information about prognostic factors in RM HNC pts receiving first-line platinum-based chemotherapy and cetuximab. Moreover, we lack survival data in a real-world population, without the selection bias affecting pts enrolled in clinical trials. Methods: We evaluated all consecutive pts treated from 1/2007 to 12/2016 in 6 Italian Centres. The following baseline prognostic factors were investigated: sex, age, site of disease, tumor grading, HPV status for oropharyngeal cancer, performance status (PS), weight loss in the previous 3 months (less/more than 5%), comorbidities (according to ACE-27), residual tumor at primary site, previous chemotherapy or cetuximab in curative setting, previous radiotherapy, platinum type (cisplatin/carboplatin, CBDCA), chemotherapy schedule (weekly/3-weekly), platinum and cetuximab doublet or with a third drug (i.e. 5FU or paclitaxel). For each potential predictor variable, Kaplan-Meier curves for OS and PFS were estimated, and a Log-rank test was used to compare survivorship in different levels of the variable. A Cox proportional hazard model was run including only predictors characterized by a significant (p < 0.05) Log-rank test. Results: We analyzed 340 pts, with a median PFS/OS of 5.0/10.6 months. The 1-year and 3-year OS rate for all pts was 44.2% (CI: 39.1-50.0) and 7.8% (CI: 5.1-12.0). Only one out of two pts received a second-line therapy. In univariate analysis lower OS was associated with PS > 0 (p < 0.001), residual tumor at primary site (p < 0.001) and CBDCA use (p = 0.012) while lower PFS was associated with paranasal sinus site (p = 0.008), PS > 0 (p = 0.001), CBDCA use (p = 0.035) and residual tumor at primary site (p < 0.001). All these predictors except for platinum type remained significant at multivariate analysis. Pts with clinical response to treatment carried a more favorable prognosis, while progressive disease as best response had a dismal median OS of 5.8 months. Conclusions: In non-selected RM HNC pts, we obtained a median PFS and OS of 5.0 and 10.6 months, very similar to 5.6 and 10.1 months reported in Extreme trial (Vermorken et al. 2008). At baseline, PS and residual tumor at primary site could be used to define pt prognosis. Legal entity responsible for the study: Paolo Bossi. Funding: Has not received any funding. Disclosure: P. Bossi, D. Ferrari, R. Depenni, G. Azzarello: Advisory board: Merck Serono. L.F. Licitra: Advisory board and research support: Merck Serono. All other authors have declared no conflicts of interest.

In carotid endarterectomy (CEA), EEG and somatosensory evoked potentials (SEPs) are the most commonly used monitoring techniques to prevent perioperative stroke. However, which of these methods is the most appropriate is not definitely established. Here we evaluated retrospectively simultaneous EEG and SEP recordings. Our CEA series was analyzed backward to identify 30 patients requiring carotid shunt. Shunting was performed in 7.1% of 420 consecutive CEA over a 20-month period. All CEAs were continuously monitored by multi-channel EEG and SEPs before, during and 20 min following carotid cross-clamping. The most reliable SEP criterion for shunting was marked amplitude reduction or disappearance of cortical components (N20, P25), usually associated or preceded by a unilateral or bilateral suppression of EEG activity. Three of the shunted patients had minor postoperative neurological deficits. Amongst the 390 non-shunted patients, 1 had a postoperative ischemic stroke and 1 one had a cerebral hemorrhage. Findings confirm that SEPs are less sensitive but more specific than EEG for the detection of cerebral ischemia in CEA. The number of shunting and major ischemic events in non-shunted patients associated to simultaneous EEG and SEP monitoring was lower than those commonly reported in the literature of CEA.

The occurrence of behavioral and psychological symptoms of dementia (BPSD) is currently recognized as an important aspect of Alzheimer's disease (AD). We evaluated the frequency and severity of BPSD with the Neuropsychiatric Inventory across the various degrees and phases of the disease in 50 consecutive AD outpatients. Apathy, aberrant motor activity, dysphoria and anxiety were the symptoms most frequently reported by the caregivers, ranging in the whole study sample from 46 to 74%. A clear trend towards increasing frequency with the severity of disease was found for delusions, hallucinations and aberrant motor activity. A major effect of the duration of the disease was found in the probability of developing hallucinations and aberrant motor activity. Apart from hallucinations, all BPSD were present starting from a mild degree of dementia. A better understanding of the global spectrum of BPSD in AD is warranted in order to improve the allocation of health resources toward the treatment of dementia.

This study evaluates the effect of direct observation at patients homes by a neurologist, on the frequency and type of behavioral disorders in Alzheimer's disease (AD) patients and on the related caregiver's reactions. Among 30 selected AD patients, 20 were positively screened for behavioral disturbances. The sampled caregivers were asked to complete the Revised Memory and Behavior Problems Checklist (RMBPC) in order to select patients showing behavioral problems in the week prior to the interview. Caregivers were asked to record in diary form the three most frequent behavioral disturbances in order to select the most appropriate time of day for a direct observation. A neurologist stayed at home of each patient for 1 hr on 7 days and monitored the patient's behavior. After direct observation both caregiver and neurologist completed the RMBPC separately. The total score of RMBPC provided by caregivers at baseline and after the neurologist's observation was compared using the t-test for paired samples. After the neurologist's direct observation, almost all AD patients showed an improvement in the occurrence of behavioral disturbances (p=0.04). The caregiver's reaction was also decreased compared to the baseline (p=0.03). Our study outlines the importance of non-pharmacological intervention in caring for AD patients whereby many of the effects of AD could be alleviated by behavioral management.

In the latest years, a debated question is whether the different clinical patterns of Alzheimer’s Disease (AD) are due to the presence of subtypes or, rather, to stages of the disease (Mayeux, 1985; Chui et al., 1985; Jorm, 1985; Mohr et al., 1990; Richte & Touchon, 1992; Yesavage et al., 1991, 1993; Joanette et al., 1994). Thus, an ever greater emphasis has been paid to the necessity of longitudinal studies (Berg et al., 1984; Jorm, 1985; Yesavage et al., 1991; Haxby et al., 1992; Richte & Touchon, 1992; Morris et al., 1993) and to detailed description of cognitive deficits in AD patients (Haxby et al., 1992).

The aim of this study was to establish the presence and the consistency of different cognitive profiles in AD patients taking into consideration the severity of mental impairment. Therefore we stratified 679 neuropsychological observations on 119 probable AD patients followed longitudinally on the basis of overall degree of cognitive impairment. To compare performance on tests with different score ranges we transformed raw test scores into coefficients; to summarize our results in terms of language versus visuo-spatial performance we computed indices of prevalent impairment of performance (IPIP) by subtracting the coefficients for constructional praxis from coefficients for language-related tests. Finally, we converted these indices into z-scores for each level of mental decline to identify patients with generalized, language (L) or visuo-spatial (V) prevalent impairment. The latter, 30% of the sample, can be detected at all stages of dementia. There was a higher percentage of males among language impaired patients (P<0.05). Approximately half of patients with L/V prevalent impairment continued to show such a focality when followed longitudinally. The groups did not differ in the annual rate of cognitive decline.

The Revised Memory and Behavior Problems Checklist (RMBPC) is a caregiver-report questionnaire for assessing major behavioral disturbances in dementia patients. Behavioral problems are of great relevance both for the clinical management and the caring of dementia patients. We assessed the RMBPC reliability in terms of interobserver reproducibility, by comparing the ratings of a close caregiver with the ratings of a neurologist directly observing 20 Alzheimer's disease patients in their family environment. The inter-rater agreements were measured by means of the Kappa coefficient and the Spearman's correlation coefficient. Our results support the use of the RMBPC to assess behavioral disturbances in dementia patients, with certain caveats as far as depression-related behaviors are concerned (r=0.60). The RMBPC can therefore be reliably used to monitor behavioral disturbances over the course of the disease as well as for clinical trials.