Search

BACKGROUND AND PURPOSE A large proportion of headache sufferers do not routinely seek medical care. App-based technologies permit the collection of real-world data over time and between countries that can help assess true burden of headache. This study used a mobile phone application to collect information on the real-world burden of self-diagnosed headache and to describe its impact on daily life in headache sufferers who do not routinely seek medical advice. METHODS This retrospective, non-interventional, cross-sectional study analysed self-reported data from users of the 'Migraine Buddy' app. The main objective was to describe self-reported characteristics of headache and migraine (triggers, duration, frequency), treatment patterns and impact on daily activity in headache sufferers from Australia, Brazil, France, Germany and Japan. Data including demographics, self-diagnosed episode type (headache/migraine), duration, potential triggers and impact on daily activity are reported. All analyses were exploratory and performed per country. RESULTS Self-reported data were collected from 60,474 users between August 2016 and August 2018. Approximately 90% of users were females; >60% were aged 24-45 years. Over one-third of users reported having two to five episodes of headache or migraine per month; impact included impaired concentration, being slower and missing work or social activities. Variations across countries were observed; within countries, episode characteristics were very similar for self-diagnosed headache versus migraine. CONCLUSIONS Headache tracking was used to describe the experience, impact and self-management approaches of migraine and headache sufferers in a real-world setting. Headache disorders present a range of important issues for patients that deserve more study and reinforce the need for better approaches to management.

International audience; Background: Primary findings from a recent study reported that magnesium supplementation significantly reduced stress in severely stressed subjects with low magnesemia, and additional vitamin B6 enhanced this effect. The mechanism by which combining magnesium and vitamin B6 leads to reduced stress in these subjects remains to be elucidated. This secondary analysis investigated the impact of magnesium and vitamin B6 supplementation and perceived stress on erythrocyte magnesium levels, as a marker of body magnesium status. Methods: This was a secondary analysis from an 8-week randomized controlled trial comparing oral magnesium (300 mg) and magnesium-vitamin B6 (300 mg + 30 mg) supplementation. Stress level and erythrocyte magnesium level at baseline, and change in erythrocyte magnesium and serum vitamin B6 levels at weeks 4 and 8, were analyzed. Results: Overall, 264 subjects were randomized to treatment and had evaluable Depression Anxiety Stress Scale scores (132 in each treatment arm). At baseline, stress scores, and mean serum magnesium, erythrocyte magnesium, and serum vitamin B6 concentrations were similar between arms. Although not significant between groups, a significant increase over time in erythrocyte magnesium levels was observed in the subgroup of subjects with low baseline erythrocyte magnesium levels (

Introduction A fixed dose combination (FDC) of ibuprofen 400 mg and caffeine 100 mg has been shown to be more effective than ibuprofen 400 mg alone for the treatment of acute postoperative dental pain in a phase III randomised controlled trial. A post hoc subgroup analysis of the primary data from an active-/placebo-controlled, double-blind, single-centre, parallel-group study was conducted in patients with moderate or severe baseline pain. Methods After dental surgery, patients with moderate or severe pain, which was determined on a 4-point verbal rating scale (‘no pain’ to ‘severe pain’), received a single dose of ibuprofen 400 mg/caffeine 100 mg FDC, ibuprofen 400 mg, caffeine 100 mg or placebo. Pain relief (PAR) and pain intensity were assessed 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 5, 6, 7 and 8 h after administration of study medication. The primary study endpoint was the time-weighted sum of PAR and pain intensity difference (PID) from pre-dose baseline, summed for all post-dose assessment times from 0 to 8 h (SPRID0–8h). Results There were 237 patients with moderate pain and 325 with severe pain at baseline. SPRID0–8h was significantly improved with the FDC versus ibuprofen, caffeine and placebo in the moderate and severe pain subgroups. Adjusted mean SPRID0–8h difference for the FDC versus ibuprofen was 18.19 (p, Plain Language Summary The non-steroidal anti-inflammatory drug (NSAID) ibuprofen is commonly used to relieve mild to moderate pain. Research suggests that combining ibuprofen with caffeine can increase the analgesic efficacy. Previously, a randomised, double-blind, placebo-controlled study showed that this ibuprofen/caffeine combination was significantly more effective than ibuprofen alone for relieving pain after dental surgery (wisdom tooth removal). Patients in that study had moderate or severe pain, so the researchers conducted another analysis of the study data to investigate how well the ibuprofen/caffeine combination worked in patients with moderate pain and in patients with severe pain. The study found that a single dose of ibuprofen/caffeine was significantly more effective than ibuprofen alone in patients with moderate pain and in those with severe pain. The analgesic effects of ibuprofen/caffeine were more marked in patients with moderate pain than in those with severe pain. This indicates that ibuprofen/caffeine is an effective pain reliever for patients with moderate pain, and to a lesser extent in patients with severe pain.

Prevalence and burden of headache disorders in real-world settings is relatively unstudied. We explored the associations between passively collected activity data, headache burden, and quality of life in headache sufferers.Data from wearable activity tracking devices and daily short questionnaires were collected over 12 weeks to assess occurrence of headache, activity, quality of life and self-rated health. Variables were analyzed using a series of mixed-effects models and stratified based on headache type. Multiple linear and logistic regressions were used to analyze treatment preferences.Behaviors inferred from activity tracker data suggested that individuals slept more, had reduced physical activity, and had lower maximum heart rate on days with headache. As headache-specific impact on quality of life increased, activity and maximum heart rate decreased and sleep increased. Headache days with higher self-rated health were associated with less napping, higher step count and maximum heart rate, correlating with increased activity. Migraineurs experienced greater burden in everyday life compared with tension-type headache sufferers.This study adds to existing evidence that activity trackers can be used to quantify headache burden in real-world settings and aid in understanding symptom management.

Introduction The aim of this study was to evaluate the efficacy and safety of a new hard-boiled lozenge formulation containing ambroxol 20 mg versus placebo for the relief of sore throat in patients with acute pharyngitis. Methods This was a phase 3, randomized, double-blind, placebo-controlled, parallel-group multicenter trial conducted between June and September 2018 in South Africa. Patients with a diagnosis of acute pharyngitis, onset ≤ 72 h, and sore throat pain of at least moderate intensity were randomized to receive either ambroxol 20 mg or placebo hard-boiled lozenges. The primary efficacy endpoint was the normalized time-weighted sum of pain intensity differences (SPID) from baseline over 3 h following administration of the first lozenge (SPIDnorm,0–3h). Secondary efficacy endpoints included SPID 24 h after the first lozenge intake (SPIDnorm,0–24h) and patient assessment of efficacy at 3 and 24 h after the first lozenge. Results Of 422 patients from 11 centers, 390 were randomized to one of the two treatment groups (n = 196, ambroxol; n = 194, placebo) and 388 were analyzed (modified intention-to-treat). The mean ± standard deviation SPIDnorm,0–3h values were −0.386 (0.259) and −0.366 (0.243) in the ambroxol and placebo groups, respectively, and the adjusted mean ± standard error SPIDnorm0–3h difference between ambroxol and placebo was −0.020 (0.025) (p = 0.443). Comparable results between treatment groups were also found for SPIDnorm,0–24h and patient assessment of efficacy at 3 and 24 h after the first lozenge. The incidence of treatment-emergent adverse events (TEAEs) was similar between treatment groups (11.7% for ambroxol versus 9.3% for placebo). Conclusion Although marked pain relief was observed over the first 3 h of treatment, superiority of ambroxol 20 mg hard-boiled lozenges versus placebo was not demonstrated in this study. Trial Registration NCT03583658. Funding Sanofi-Aventis Group.

Background With headache experienced by up to 75% of adults worldwide in the last year, primary headache disorders constitute a major public health problem, yet they remain under-diagnosed and under-treated. Headache prevalence and burden is changing as society evolves, with headache now occurring earlier in life. Contributing factors, mostly associated with changing life style, such as stress, bad posture, physical inactivity, sleep disturbance, poor diet and excess use of digital technology may be associated with the phenomenon that could be labelled as ‘21st century headache’. This is especially notable in workplace and learning environments where headache impacts mental clarity and therefore cognitive performance. The headache-related impact on productivity and absenteeism negatively influences an individual’s behaviour and quality of life, and is also associated with a high economic cost. Since the majority of sufferers opt to self-treat rather than seek medical advice, substantial knowledge on headache prevalence, causation and burden is unknown globally. Mapping the entire population of headache sufferers can close this knowledge gap, leading to better headache management. The broad use of digital technology to gather real world data on headache triggers, burden and management strategies, in self-treated population will allow these sufferers to access appropriate support and medication, and therefore improve quality of life. Conclusion These data can yield important insights into a substantial global healthcare issue and form the basis for improved patient awareness, professional education, clinical study design and drug development.

Background: Unexplained fatigue is a common complaint. When underlying disease causes have been eliminated, lifestyle measures and supplementation can be indicated. Elaborating on clinical findings that G115®, a dry extract from the root of Panax ginseng, combined with vitamins and minerals could alleviate fatigue, this open label study aimed at assessing its effect on perceived fatigue and energy. Methods: Healthy adults self-reporting fatigue (n = 103) completed the Multidimensional Fatigue Inventory questionnaire. They rated their perceptions of mental and physical fatigue, energy, performance, and stress at baseline and 15, 30, 60 and 90 days after a daily intake of 40 mg G115® formulated with vitamins and minerals. Results: Compared with baseline values, mean self-perception of general fatigue was reduced by −7.55 units [95% CI: −8.44; −6.66] (−41.8%, p < 0.0001) at 90 days. All assessed perception ratings (mental and physical fatigue, reduced activity and motivation, performance, and stress) were significantly and steadily improved from two weeks after supplementation up to study’s end. Overall satisfaction with the ability of the product to reduce fatigue reached 85% at Day 90. Conclusion: Daily intake with G115® extract formulated with vitamins and minerals suggests an improvement of self-perception of fatigue and energy in a fatigued adult population.

We examined factors that may impact cabazitaxel treatment duration in a real-life setting in a compassionate use program, expanded access program, and prospective observational study in metastatic castration-resistant prostate cancer (mCRPC). Patients with mCRPC previously treated with docetaxel (N = 1,621) received cabazitaxel 25 mg/m2 intravenously every 3 weeks until disease progression, death, unacceptable toxicity or physician/patient decision. The median number of cabazitaxel cycles was six (range, 1&ndash, 49), 708 patients (43.7%) received &gt, 6 cycles. Patients receiving &gt, 6 cycles tended to have a better Eastern Cooperative Oncology Group performance status of 0&ndash, 1 (p = 0.0017 for &le, 6 vs. &gt, 6 cycles). Overall, 348 patients (21.5%) were &ge, 75 years of age, 139 (39.9%) received &gt, 6 cycles. The main reason for discontinuation was disease progression, however, in patients receiving 1&ndash, 2 cycles, the main reason for discontinuation was adverse events. Only 52 patients (3.2%) progressed during cycles 1&ndash, 2. Cabazitaxel was well tolerated in these studies, which included some elderly and frail patients, offering clinicians an important treatment option in the management of mCRPC. Proactive management of adverse events may allow patients to receive a higher number of cabazitaxel cycles and derive greater benefit.

Càncer de pròstata resistent a la castració metastàsica; Cabazitaxel; Qualitat de vida relacionada amb la salut Cáncer de próstata metastásico resistente a la castración; Cabazitaxel; Calidad de vida relacionada con la salud Metastatic castration-resistant prostate cancer; Cabazitaxel; Health-related quality of life Objectives To obtain routine clinical practice data on cabazitaxel usage patterns for patients with metastatic castration‐resistant prostate cancer (mCRPC) and to describe physician‐assessed cabazitaxel effectiveness, health‐related quality of life (HRQoL) and safety. Patients and Methods CAPRISTANA was an international, observational cohort study examining cabazitaxel use for the treatment of patients with mCRPC. Effectiveness was assessed by overall survival (OS), progression‐free survival (PFS), time to treatment failure (TTF) and disease control rate. HRQoL was assessed using the Functional Assessment of Cancer Therapy‐Prostate questionnaire (FACT‐P) and the three‐level European Quality of Life questionnaire (EQ‐5D‐3L). Safety was assessed by adverse event (AE) reporting. Results A total of 189 patients were treated across 54 centres between April 2012 and June 2016. At baseline, 58.7% had ≥1 comorbidity, 93.7% had an Eastern Cooperative Oncology Group performance status ≤1, and 60.1% had a Gleason score at diagnosis of ≥8. Patients received a median of 6 cabazitaxel cycles; 84.7% received cabazitaxel as second‐line therapy. The median OS, PFS and TTF were 13.2, 5.6 and 4.4 months, respectively. Cabazitaxel led to disease control in 52.9% of patients. HRQoL was maintained (40.3%) or improved (32.2%) in 72.5% of patients based on total FACT‐P scores. Interestingly, 53.6% of patients reported pain improvement and a further 21.2% maintained pain control based on FACT‐P prostate cancer‐specific pain scores. The most common treatment‐related grade ≥3 AEs were neutropenia (7.9%) and anaemia (2.1%). Conclusion Patients in CAPRISTANA treated with cabazitaxel had similar disease outcomes and safety profiles compared with large phase III clinical trials. Most patients had maintained or improved HRQoL scores; >70% of patients had maintained or improved pain control. Research and analysis was supported by Sanofi. The authors were responsible for all content and editorial decisions, and received no honoraria for development of this manuscript. Editorial support was provided by Amber Wood and Anna Longjaloux of MediTech Media, funded by Sanofi. The authors would also like to thank Teri Michelini of Sanofi for support in preparing this manuscript.

Purpose There is a major clinical need to devise an optimal treatment sequence for the multiple therapy options available for patients with metastatic castration-resistant prostate cancer (mCRPC). In the absence of prospective clinical trials, sequencing information can be derived from large, real-world registry studies. Patients and Methods PROXIMA (Treatment Patterns in Patients With Metastatic Castration-Resistant Prostate Cancer Previously Treated With Docetaxel-Based Chemotherapy) is a large, global, prospective registry study evaluating real-world treatment patterns of patients with mCRPC who experience disease progression during or after docetaxel therapy. Patients were enrolled worldwide between 2011 and 2014. Treatments were determined by the treating physicians and recorded in categories of chemotherapy, hormonal therapy, targeted therapy, immunotherapy, and palliative therapy. Treatment sequencing patterns, response to treatment, and types of progression were recorded and analyzed. Progression-free survival and overall survival with different treatment modalities were analyzed using Kaplan–Meier method. Results Treatment patterns were evaluated in 903 patients. Therapy selection was influenced by region. Hormonal therapy (57.5%) and taxane chemotherapy (26.4%) were the most frequently administered first subsequent treatments after docetaxel. Tumor responses to first subsequent treatment were observed in 22.6% of evaluable patients. Overall survival and progression-free survival did not differ significantly across different treatment modalities. Conclusion Identifying an optimal treatment sequence is vital for improving the care of patients with mCRPC. The PROXIMA registry provided a representative sample of global data on real-world treatment patterns for patients with mCRPC previously treated with docetaxel. These data can be used to devise optimal therapy sequences and inform treatment decisions.

IntroductionEvidence from a recent randomised controlled trial1 suggests that in severely stressed subjects with low magnesemia, supplementation with magnesium (Mg) in combination with vitamin B6 (B6) provides greater benefits than Mg alone. B6 was reported to facilitate Mg absorption and its cellular uptake and to exert synergistic effect with Mg. The current secondary analysis explored the relationship between Mg-B6 combination and erythrocyte Mg concentration, used as a biomarker of body Mg status.Material and MethodsAn 8-week, Phase IV, controlled, single-blinded, parallel-group trial (EudraCT Number 2015-003749-24) stratified by sex was conducted in adults (n = 264) with a Depression Anxiety Stress Scales - stress subscale score (DASS-42SS) > 18 and serum Mg of 0.5–0.85mmol/L, randomised 1:1 to daily oral Mg-B6 (Magne B6®, Mg 300 mg; B6 30mg) or oral Mg alone (Magnespasmyl®, Mg 300mg). Outcomes were stress score, serum Mg (mmol/L), erythrocytes Mg (mmol/L), and serum B6 (nmol/L) from baseline to Week4 and Week8. Data are given as mean(SD) values.Results & DiscussionBaseline characteristics. Baseline magnesemia was 0.80(0.04) for both groups. Erythrocyte Mg concentration for the lower quintile of the studied population (n = 53) was 0.73–1.62, below the normal range of 1.65–2.65 in general population. The mean stress score in this subgroup was higher [29.5(6.3)], but not significantly different from that in other quintiles [lowest value: 26.1(7.6)]. Baseline B6 serum level for the lower B6 quintile (5–23), below the normal range for general population, was suggestive of possible B6 deficiency.Treatment effects. Both treatments increased slightly but not significantly erythrocyte Mg level from baseline to Week8 [1.84(0.03) to 1.86(0.03), and 1.86(0.03) to 188(0.03), respectively for Mg + B6 and Mg groups]. Significant changes were observed in subjects with low erythrocyte Mg level at baseline (< 1.6), namely an increase of + 0.13(0.04–0.22) for Mg + B6 and + 0.17(0.08–0.25) for Mg groups, but with no difference between treatments. Moreover, Mg + B6 supplementation led to a significant change (p < 0.0001) in serum B6 level between baseline and Wk8 [55.9(4.8) to 370.9(11.3)], as compared to Mg alone [51.9(4.8) to 51.5(11.3)].In conclusion, both treatments significantly increased erythrocyte Mg in subjects with low Mg status. B6 supplementation did not lead to further increase in erythrocyte Mg level. We have previously shown that severely stressed population benefits from B6 supplementation in combination with Mg, however, the mechanism of the interaction between Mg and B6 remains to be elucidated.1Sponsored by Sanofi

// Zafar Malik 1 , Axel Heidenreich 1 , 2 , Sergio Bracarda 3 , Alexandros Ardavanis 4 , Philip Parente 5 , Hans-Joerg Scholz 6 , Ayse Ozatilgan 7 , Evelyne Ecstein-Fraisse 8 , Simon Hitier 9 and Giuseppe Di Lorenzo 10 1 The Clatterbridge Cancer Centre NHS Foundation Trust, Wirral, UK 2 Department of Urology, Uro-Oncology, Robot-Assisted and Specialized Urologic Surgery, University Hospital Cologne, Cologne, Germany 3 Azienda USL Toscana Sud-Est, Istituto Toscana Tumori (ITT), Ospedale San Donato, Arezzo, Italy 4 Oncology Hospital AGIOS SAVVAS Oncology Clinic, Athens, Greece 5 ECRU-Oncology, Victoria, Australia 6 Asklepios Klink GmbH Weissenfels, Weissenfels, Germany 7 Sanofi, Cambridge, Massachusetts, USA 8 Sanofi, Paris, France 9 Sanofi, Chilly-Mazarin, France 10 Department of Clinical Medicine and Surgery, University of Naples Federico II, Napoli, Italy Correspondence to: Zafar Malik, email: zaf.malik@nhs.net Keywords: mCRPC; cabazitaxel; CUP; EAP; real-world Received: January 08, 2019 Accepted: April 29, 2019 Published: June 25, 2019 ABSTRACT Background Cabazitaxel is a second-generation taxane approved for use in patients with metastatic castration-resistant prostate cancer (mCRPC) previously treated with docetaxel. Early access programmes were established to allow eligible patients with mCRPC access to cabazitaxel before regulatory approval. Materials and Methods The primary objective was to allow access to cabazitaxel before commercial availability for patients with mCRPC whose disease had progressed during or after chemotherapy with docetaxel; the secondary objective was overall safety. Patients received cabazitaxel 25 mg/m 2 on Day 1 of a 21-day cycle, with daily oral 10 mg prednisone/prednisolone. G-CSF was administered per ASCO guidelines. Results In total, 1432 patients received cabazitaxel across 41 countries between 2010 and 2014 (median 6.0 treatment cycles [range 1–49]). The most frequently occurring treatment-emergent adverse events (TEAEs) possibly related to treatment were diarrhoea (33.3%), fatigue (25.4%) and anaemia (23.7%); the most frequently occurring possibly related Grade 3/4 TEAEs were neutropenia (18.7%) and febrile neutropenia (6.9%). G-CSF was administered in ≥ 1 cycle in 64% of patients (10.1% therapeutic use; 57.8% prophylactic use; 9.7% both uses). Conclusion The safety profile of cabazitaxel in this pooled analysis of two cabazitaxel early access programmes was manageable and consistent with previous Phase III trials (TROPIC, PROSELICA).

Meta-analysis is the statistical methodology to combine study results from different clinical trials to address questions of clinical interest. It is done as a systematic review of the relevant clinical trials. Meta-analysis is usually conducted in three steps: data/study selection, statistical analysis, and interpretation. We discuss the practical issues in these three steps and introduce how a proper meta-analysis is conducted. A case study of meta-analysis of fatal adverse events in cancer patients treated with aflibercept is introduced and its limitations are discussed.

Background Docetaxel/cisplatin/infusional 5-fluorouracil (5-FU; DCF) is a standard chemotherapy regimen for patients with advanced gastric cancer (GC). This phase II study evaluated docetaxel/oxaliplatin (TE), docetaxel/oxaliplatin/5-FU (TEF), and docetaxel/oxaliplatin/capecitabine (TEX) in patients with advanced GC. Patients and methods Patients with metastatic or locally recurrent gastric adenocarcinoma (including carcinoma of the gastro-oesophageal junction) were randomly assigned (1 : 1 : 1) to TE, TEF, or TEX. Each regimen was tested at two doses before full evaluation at optimized dose levels. The primary end point was progression-free survival (PFS). Overall survival (OS), tumour response, and safety were also assessed. A therapeutic index (median PFS relative to the incidence of febrile neutropenia) was calculated for each regimen and compared with DCF (historical data). Results Overall, 248 patients were randomly assigned to receive optimized dose treatment. Median PFS was longer with TEF (7.66 [95% confidence interval (CI): 6.97–9.40] months) versus TE (4.50 [3.68–5.32] months) and TEX (5.55 [4.30–6.37] months). Median OS was 14.59 (95% CI: 11.70–21.78) months for TEF versus 8.97 (7.79–10.87) months for TE and 11.30 (8.08–14.03) months for TEX. The rate of tumour response (complete or partial) was 46.6% (95% CI 35.9–57.5) for TEF versus 23.1% (14.3–34.0) for TE and 25.6% (16.6–36.4) for TEX. The frequency and type of adverse events (AEs) were similar across the three arms. Common grade 3/4 AEs were fatigue (21%), sensory neuropathy (14%), and diarrhoea (13%). Febrile neutropenia was reported in 2% (TEF), 14% (TE), and 9% (TEX) of patients. The therapeutic index was improved with TEF versus TEX, TE, or DCF. Conclusion These results suggest that TEF is worthy of evaluation as an arm in a phase III trial or as a backbone regimen for new targeted agents in advanced GC. ClinicalTrials.gov Identifier Trial registration number: NCT00382720.

Aims A better understanding of treatment patterns and outcomes in different countries should improve the management of patients with gastric cancer globally. The REgistry of GAstric Cancer Treatment Evaluation (REGATE) study was established to evaluate variations in gastric cancer disease characteristics and treatment patterns in different parts of the world. Methods REGATE was a prospective international registry enrolling patients with newly diagnosed gastric cancer at any stage of the disease. Results A total of 10 299 patients (65% male; mean age 59 years) were recruited in 22 countries between 2004 and 2008. Tumor location at a proximal site was more common in Europe, Latin America and North Africa (approximately 20%) than in Asia-Pacific, where antral location predominated. Signet-ring cell histology predominated except in Europe, where adenocarcinoma was most prevalent. Stage I cancers were more frequent in Asia-Pacific (39%) versus other regions (6−18%), whereas stage IV cancers were more frequent outside Asia-Pacific. Surgery was planned for most patients, although in general fewer patients actually received surgery than originally planned. Adjuvant therapy and palliative care were generally used more frequently than originally planned. Overall, 15% of patients received no treatment (Asia-Pacific 8%; Indian subcontinent 25%). Conclusions These results provide a comprehensive database representative of gastric cancer disease characteristics and treatment patterns across the world.

Aims The objective of the registry of gastric cancer treatment evaluation (REGATE) study was to evaluate approaches to gastric cancer treatment in different geographical regions. Methods REGATE enrolled patients with newly diagnosed gastric cancer at any stage of the disease from the Asia–Pacific region, Europe, the Indian subcontinent, Latin America and North Africa between 2004 and 2008. Results Among 9965 patients, 69% received surgery, 40% palliative care, 29% adjuvant therapy and 2% neoadjuvant therapy; 15% received no treatment. Combination treatment (mostly surgery/adjuvant) was used in one-third of patients. Overall, 90% received chemotherapy (mostly fluoropyrimidine/platinum combinations but with marked geographical variation) and 21% received radiotherapy. Curative surgery alone was used most frequently for stages 0–II cancers and was employed more often in Europe (55%) and the Asia–Pacific (48%) than in other regions (27–35%). Asia–Pacific and Indian subcontinent patients were more likely to have a distal subtotal gastrectomy and less likely to undergo total gastrectomy than patients in other regions. Lymph node D2 dissection was favored in the Asia–Pacific, Europe and Latin America, whereas D1 dissection was used more in the Indian subcontinent and North Africa. Conclusion These data showing geographical differences in the approaches to gastric cancer treatment may promote the optimization of the management of gastric cancer globally.

Macrophage elastase (recombinant human matrix metalloproteinase-12, rhMMP-12), was instilled in mouse airways, inducing an early inflammatory response characterized by neutrophil recruitment and cytokine release in the bronchoalveolar lavage (BAL) fluids, followed by a delayed macrophage recruitment. We investigated the role played by alveolar macrophages and neutrophils in the delayed macrophage influx induced by rhMMP-12 (8 x 10(-3) U/mouse) in A/J mice. Mice depleted of circulating neutrophils, using a cytotoxic antibody, did not present an increase in neutrophil numbers in bronchoalveolar lavage fluids, 4 h and 24 h after rhMMP-12 instillation but the macrophage recruitment was not modified as compared to control mice at 7 days. Similar results were obtained using mice when the gene for neutrophil elastase was knocked out. Intranasal instillation of clodronate liposomes, 72 h prior to rhMMP-12 instillation, induced macrophage depletion which did not modify the macrophage recruitment at 7 days. Moreover, the stimulation of mouse macrophages by rhMMP-12 did not elicit the release of cytokines in culture supernatants. These results indicate that resident alveolar macrophages and recruited neutrophils do not play a role in the delayed macrophage recruitment induced by rhMMP-12.

Direct instillation of a recombinant human form of MMP-12 (rhMMP-12) in mice airways elicited an early inflammatory response characterized by neutrophil influx, cytokine release and gelatinase activation followed by a delayed response, mainly characterized by macrophage recruitment. As this experimental model of lung inflammation partially mimics some features of chronic obstructive pulmonary disease (COPD), we have investigated the effects of treatment by anti-inflammatory compounds, dexamethasone and rolipram and a non-specific matrix metalloproteinase (MMP) inhibitor, marimastat. The compounds were administrated orally, 1 h before rhMMP-12 instillation (8 x 10(-3) U/mouse). Total and differential cell counts were evaluated in the bronchoalveolar lavage fluids. Cytokines and MMP-9 were quantified in bronchoalveolar lavage fluids and in lung homogenate supernatants. Marimastat (100 mg/kg), dexamethasone (10 mg/kg) and rolipram (0.1 and 0.3 mg/kg) were able to decrease significantly neutrophil recruitment at 4 and 24 h after rhMMP-12 instillation, but only marimastat (30 and 100 mg/kg) was effective at decreasing the macrophage recruitment occurring at day 7. Marimastat (100 mg/kg), dexamethasone (10 mg/kg) and rolipram (0.3 mg/kg) reduced significantly IL-6, KC/CXCL1, MIP-1alpha/CCL3 and MMP-9 levels in bronchoalveolar lavage fluid. Similar results were obtained in lung homogenates except with rolipram. Dexamethasone and rolipram were able to inhibit the early inflammatory response but were ineffective to limit the macrophage influx. In contrast, marimastat was able to reduce early and late response. These data indicate that MMP-12 instillation in mice could highlight some of the inflammatory response seen in COPD and could be used for the pharmacological evaluation of new anti-inflammatory mechanisms of action.

To compare the efficacy and tolerability of taxane and nontaxane therapy in senior adults with chemonaïve metastatic castration-resistant prostate cancer (mCRPC), and examine the effect of patient health status on outcomes.Between 2009 and 2011, 333 patients aged≥70 years with mCRPC were enrolled in a prospective international registry. Patients were categorized as having received taxane-based or nontaxane therapy, and classified as fit, vulnerable, frail, or terminal, according to investigator judgement or International Society of Geriatric Oncology guidelines. Efficacy measures included overall survival (OS) and progression-free survival. Grade 3/4 toxicities were recorded. Predictors of OS were identified using multivariate Cox regression.The proportions of fit/vulnerable/frail patients were 65%/14%/17% (International Society of Geriatric Oncology), and 39%/43%/17% (investigator). In single-factor analyses, taxane therapy improved OS (hazard ratio [95%CI] = 0.53 [0.30-0.93]; P = 0.027) and progression-free survival (hazard ratio [95% CI] = 0.55 [0.40-0.76]; P0.001) vs. nontaxane therapy. Patients with frailty also benefited from taxane therapy (adapted regimen in 52%). In multivariate analysis, taxanes improved OS even with poor prognostic factors present (P = 0.017); age was unrelated to prognosis. Taxane therapy was well tolerated; most common grade 3/4 toxicities (taxane vs. nontaxane) were fatigue (17% vs. 4%), nausea/vomiting (14% vs. 5%) and neutropenia (10% vs. 1%).The results of this nonrandomized, observational study suggest that first-line taxane therapy may benefit senior adults with mCRPC more than alternative therapies. Treatment decisions should not be based on chronological age.

Macrophage elastase (MMP-12) is a metalloproteinase able to degrade extracellular matrix components such as elastin. As many MMPs, MMP-12 is involved in acute and chronic lung injury. However, its role in the inflammatory process of the lung parenchyma is not clearly understood. In this study, we have investigated the effects of airway instillation of rhMMP-12 on inflammatory cell recruitment, cytokine release and gelatinase expression in bronchoalveolar lavage fluid (BALF) or in lung homogenate supernatants in mice. Numbers of total and individual cell types were examined in BALF during the first 72 h following rhMMP-12 instillation. A marked recruitment of neutrophils was observed with a maximum increase at 18 h. This cellular recruitment was associated with a very transient increase in IL-6, TNF-α MIP-1α, MCP-1 and KC levels and gelatinase expression in BALF and in lung homogenate supernatants. From days 4 to 15, performing the same analyses, we observed an important and stable recruitment of macrophages in BALF in absence of the other studied inflammatory markers. These results demonstrate that rhMMP-12 itself is able to induce an early inflammatory response characterized by neutrophil infiltration, cytokine release and gelatinase activation followed by a later response composed mainly of macrophage recruitment.

We read with interest ‘Treatment-related mortality with aflibercept in cancer patients” by Wei-Xiang Qi and colleagues in the April 2014 issue of your journal [1], in which they conclude that the use of aflibercept is associated with an increased risk of fatal adverse events (FAEs) compared with controls. It is our opinion that this meta-analysis has some important methodological flaws. The analysis was conducted using the published data for 3,060 patients, with a range of solid tumors. In 954 patients who received single-agent aflibercept and had data available for analysis, the authors estimated the overall incidence of FAEs to be 5.1 % (95 % CI 3.8–6.8) [1]. However, the authors erroneously included the VENICE trial [2] in their estimate of the incidence of FAEs with single-agent aflibercept. The phase III VENICE trial compared docetaxel and prednisone plus either aflibercept or placebo for the treatment of metastatic castration-resistant prostate cancer and therefore is not informative for single-agent aflibercept. Also, in the VENICE trial, the aflibercept dose was 6 mg/kg and not the 4 mg/kg dose approved for use in patients with metastatic colorectal cancer (mCRC) who have progressed on prior oxaliplatin-containing therapy. Another important point is that the methodology used for calculating the incidence of FAEs may be inappropriate. By using the logit transformation (extremely unusual when estimating a single proportion), the authors implicitly assigned weights that were not proportional to the number of patients. Consequently, studies with high FAE rates are over-weighted. The exclusion of the VENICE trial and the subsequent assignment of appropriate weights result in an estimate of the incidence of FAEs of 3.2 % (95% CI 1.6–5.6) for single-agent aflibercept. Also, in an attempt to determine the specific contribution of aflibercept to the development of FAEs, the authors calculated the odds ratio (OR) for FAEs from four randomized clinical trials comparing aflibercept with controls in 2,715 patients. Most surprisingly, they excluded from their analysis the pivotal, randomized, VELOUR trial, in patients with mCRC who were resistant to or had progressed on a prior oxaliplatin-containing regimen [3], which led to the only approved indication for aflibercept. Furthermore, the authors extracted incorrect FAE values for the phase III VITAL study [4]. The correct values for FAEs in the VITAL study [4] are 32 FAEs in the aflibercept group versus 18 in the control group, and not 11 versus 1. For all FAE recalculations that included the VITAL study, we used the data from the referenced publication. Inclusion of the VELOUR study in the estimates of FAEs results in an OR of 1.516 (95 % CI 1.124–2.053, one-sided p = 0.0028). Taking into account only the studies using the approved 4 mg/kg dose and including the pivotal VELOUR study [3], the OR is 1.289 (95 % CI 0.836–2.001, one-sided p = 0.1354), corresponding to an increased risk of FAEs of approximately 29 %. For the two randomized controlled trials using the higher 6 mg/kg dose of aflibercept (2,137 patients) [2, 4], the OR is 1.767 (95 % CI 1.156–2.734, one-sided p = 0.0037), corresponding to an increase in the relative risk of FAEs of 77 %. We would like to reiterate that in this meta-analysis, the studies used to estimate the incidence of FAEs with aflibercept in cancer patients are all outside the labeled indication for aflibercept. Also, the exclusion of the VELOUR trial renders the results of this meta-analysis incomplete and prone to misinterpretation, although the authors conclude that the use of aflibercept remains justified in its approved indication in patients with mCRC. Importantly, and as reported previously for VELOUR [3] and other clinical studies [5], the safety profile associated with combining aflibercept with chemotherapy is similar to that seen with other anti-VEGF agents, with no new toxicity signals observed.

A better understanding of treatment patterns and outcomes in different countries should improve the management of patients with gastric cancer globally. The REgistry of GAstric Cancer Treatment Evaluation (REGATE) study was established to evaluate variations in gastric cancer disease characteristics and treatment patterns in different parts of the world.REGATE was a prospective international registry enrolling patients with newly diagnosed gastric cancer at any stage of the disease.A total of 10 299 patients (65% male; mean age 59 years) were recruited in 22 countries between 2004 and 2008. Tumor location at a proximal site was more common in Europe, Latin America and North Africa (approximately 20%) than in Asia-Pacific, where antral location predominated. Signet-ring cell histology predominated except in Europe, where adenocarcinoma was most prevalent. Stage I cancers were more frequent in Asia-Pacific (39%) versus other regions (6-18%), whereas stage IV cancers were more frequent outside Asia-Pacific. Surgery was planned for most patients, although in general fewer patients actually received surgery than originally planned. Adjuvant therapy and palliative care were generally used more frequently than originally planned. Overall, 15% of patients received no treatment (Asia-Pacific 8%; Indian subcontinent 25%).These results provide a comprehensive database representative of gastric cancer disease characteristics and treatment patterns across the world.

Various authors have proposed approximations to permutation tests of independence between two data tables. We develop approximations based on explicit expressions of the first three moments of three different test statistics under the permutation distribution. The rejection level is then determined by using a Pearson-type III distribution matching the values of the first three moments. We present three examples in which the relative merits of the test statistics are examined and the results of the approximation procedure are compared with explicit permutation tests.

Introduction The Phase III TROPIC study (NCT00417079) showed that CbzP provides a survival advantage vs mitoxantrone + P in pts with mCRPC previously treated with D (hazard ratio [HR] 0.70; P Methods Total enrolment is expected to be ∼1600 pts with mCRPC from 250 centres worldwide. Pts will receive CbzP (25 mg/m Q3W + 10 mg P oral QD) until progressive disease (PD), death, unacceptable toxicity, physician/pt decision or commercial availability of Cbz. Pts will be followed for 30 days after last administration for safety. Use of G-CSF is recommended as per ASCO guidance. Results Baseline and safety data from the first 919 pts enrolled in the 2 programmes in 30 countries are shown. Mean age was 68 yrs with 21.2% ≥ 75 yrs. All pts had ECOG ≤ 2; the most common sites of metastases were bone (91.7%), lymph nodes (regional 29.9% and distant 28.0%), lung (11.2%) and liver (9.9%). 16.8% had PD during their last D regimen. The median number of treatment cycles was 4 (range 1–16); median relative dose intensity (all cycles) was 99%. Among the 502 pts who stopped CbzP, the most common discontinuation reasons were PD (40.9%), adverse event (AE; 26.9%) and commercial availability of Cbz (15.8%). G-CSF was administered to 60.5% of pts (prophylactic and/or therapeutic use). Overall, 40.4% of pts had Grade 3–4 AEs possibly related to CbzP, the most common relevant being febrile neutropenia (FN; 6.1%), fatigue (3.6%), diarrhoea (3.0%) and nausea (1.0%). Of the 41 pts (4.5%) with AEs leading to death, 21 (2.3%) had at least 1 AE possibly related to CbzP. Conclusion These interim results provide additional safety data on CbzP use in the routine clinical practice setting across the globe. Treatment with CbzP was clinically manageable and FN rate was lower than in the TROPIC study, possibly due to the use of G-CSF as prophylaxis. Disclosure Z.I. Malik: Has participated in an advisory board for, and has received honoraria and funds to attend conferences from Sanofi, G. Di Lorenzo: Has acted as a compensated consultant for Sotio and TEVA, P. Parente: Has participated in an advisory board for, and has received research funding from, Sanofi, W.R. Gerritsen: Has acted as a compensated consultant for Algaia Biomedical Ventures, S. Hitier: Is a Sanofi employee (biostatistician) and owns shares in Sanofi, A. Heidenreich: Has participated in advisory boards for Sanofi, Janssen Cilag and Astellas. Has received honoraria from, Sanofi, Astellas, Amgen and Janssen Cilag, A. Bahl: Has participated in an advisory board for, and received research funding from Sanofi. All other authors have declared no conflicts of interest.

(AE) in 21.7%. Dose reductions (17.3%) and delays (36.4%) were due to related AEs in 15.3% and 15.7%, respectively. During the study, 248 pts (55.0%) received G-CSF, 214 (47.5%) at Cycle 1. In pts with prophylactic G-CSF use at Cycle 1 (n = 137, 30.4%), neutropenia and febrile neutropenia occurred in 4.4% and 0.7%, respectively; in pts without G-CSF at Cycle 1, respective rates were 7.6% and 1.7%. Treatment-emergent AEs (TEAEs) occurred in 83.4% (Grade 3/4 51.0%) and TEAEs related to treatment in 72.9% (Grade 3/4 41.2%). Most frequent Grade 3/4 TEAEs related to Cbz + P were neutropenia (16.9%), febrile neutropenia (8.9%), anaemia (6.0%), leukopenia (5.1%) and fatigue (4.0%). 30 deaths (6.7%) occurred, due to progression, AE or other reasons during the on-treatment period (first dose – 30 days after last dose), or due to possibly related AEs during follow up. Conclusions: In pts with prior taxane exposure, Cbz + P had a predictable, manageable safety profile consistent with the TROPIC trial. A low rate of neutropenic complications in pts with prophylactic G-CSF at Cycle 1 supports use of G-CSF to prevent haematological AEs in pts at risk. Disclosure: A. Heidenreich: has provided a consultancy role and been a member of advisory boards for Astellas, Bayer, Janssen-Cilag, Sanofi Aventis, TEVA, and Dendreon, and has received research funding from Astellas and Sanofi Aventis;H. Ozen: has been a member of advisory boards for Sanofi, Janssen and Astellas;I.M. van Oort: has been a member of advisory boards for Sanofi, Astellas, Janssen and Bayer;W. R. Gerritsen: D has been a member of advisory boards for Sanofi;E. Efstathiou: has provided a consultancy role, received honoraria and been a member of advisory boards for Johnson and Johnson, Sanofi, Millennium/Takeda and Bayer; J. Lee: has received research funding from Bayer; A. Boumessous: is an employee of Sanofi; Z. Su: is an employee and stock holder of Sanofi.S. Hitier: is an employee and stock holder of Sanofi. All other authors have declared no conflicts of interest.

242 Background: Cbz is an established second-line treatment for pts with mCRPC, having demonstrated an overall survival benefit (with P) vs mitoxantrone with P in the Phase III TROPIC trial. The global CUP (CABAZ_C_05005) and EAP (NCT01254279) (both funded by Sanofi) aimed to provide access to Cbz + P before commercial availability and to document safety in a real-world population. We present a regional analysis of interim data from the CUP and EAP. Methods: Expected total enrolment is 1,450 pts. Pts received Cbz 25 mg/m2 IV Q3W + P 10 mg QD until disease progression (PD), death, unacceptable toxicity or physician/pt decision. Granulocyte colony-stimulating factor (G-CSF) was administered per ASCO guidelines. The cut-off date for this analysis was 30 May 2012. Results: Globally, 1,301 pts have been included from eight regions (Eastern Europe [EE], Western Europe [WE], Northern Europe [NE] and Southern Europe [SE]; Asia [As]; Canada [Ca]; Latin America [LA]; and Australia [Aus]). Key data by region are shown in the Table. Median age varied from 65 years (Ca) to 70 years (Aus, LA, SE). In pts who progressed after last D line, time from last D dose to last PD ranged from 2.3 months (EE, LA) to 6.6 months (Ca). G-CSF use (therapeutic or prophylactic) at Cycle 1 varied widely between regions, from 23% (WE) to 79% (As). Between 11% (LA) and 30% (NE) of patients received ≥10 cycles of Cbz. Conclusions: In this extensive, real-world CUP/EAP cohort of mCRPC pts, substantial regional variation was observed for baseline and on-treatment parameters, including time from last D dose to last PD and G-CSF use. The low proportion of patients receiving ≥10 cycles in LA, Ca and WE may result from a limit of 10 cycles of therapy in some centers. Clinical trial information: NCT01254279. [Table: see text]

109 Background: Cbz + P demonstrated an overall survival benefit vs mitoxantrone + P in pts with mCRPC in the Phase III TROPIC trial. The CUP (CABAZ_C_05005) and EAP (NCT01254279) (both funded by Sanofi) were established to allow access to Cbz ahead of commercial availability. The programs are also evaluating Cbz safety in a real-world population. Data analyzed by age group (≤75 and >75 years) are presented here. Methods: Expected enrolment across both programs is 1,450 pts from 236 centres worldwide. Pts received Cbz 25 mg/m2 IV Q3W + P 10 mg QD until disease progression, death, unacceptable toxicity or physician/pt decision. G-CSF is administered as per ASCO guidelines. Results: As of May 30, 2012, 1,301 pts have enrolled (≤75 years: 1,061 pts [81.6%]; >75 years: 240 pts [18.4%]). Eastern Cooperative Oncology Group performance status and incidence of visceral metastases were generally balanced between treatment groups. The most frequent reasons for discontinuation were disease progression (46.8%) followed by adverse events (AEs; 24.4%) in pts ≤75 years, and AEs (36.4%) followed by disease progression (31.1%) in pts >75 years. Time from initial diagnosis to inclusion was greater in pts >75 years (median 79.66 months) than in pts ≤75 years (median 53.94 months), but time from mCRPC diagnosis to inclusion was approximately equivalent (>75 years: median 22.6 months; ≤75 years: median 20.94 months). G-CSF use was more frequent in pts >75 years (cycle 1: 62.9% of pts) compared with pts ≤75 years (cycle 1: 52.2% of pts). AEs of clinical concern were more frequent in the older age group (Grade ≥3 AEs: >75 years 64.2%; ≤75 years 54.8%). Grade ≥3 neutropenia was observed in 25.8% of pts >75 years and in 17.0% of pts ≤75 years. Conclusions: We observed several differences between age groups in baseline and on-treatment parameters, suggesting differences in the natural history of mCRPC (faster disease progression in pts ≤75 years than in pts >75 years) and secondary to treatment (AEs more frequent in pts >75 years compared with pts ≤75 years). Clinical trial information: NCT01254279.

Blue tits, as many other species, show much geographic variation in their songs. In blue tits, songs that include a trill (series of rapidly repeated notes) are widespread on the European mainland, but have not been reported in North Africa or the Canary Islands. We studied song structure of two blue tit populations from the French Mediterranean that are exposed to large differences in local selection pressures (food, vegetation, social factors). Songs from populations on the European mainland and the island of Corsica differed much in syntax, tempo, frequency, and the incidence of trills. The observed population differences could potentially be caused by a combination of different factors, including biotic and abiotic habitat characteristics, body size, and cultural drift. We suggest that songs without trill are acoustically adapted for transmission in habitats where the local breeding density of great tits is low. In addition, we suggest that songs with trill may reflect a character shift, to minimise territorial interactions with great tits.

5055 Background: Cbz + P provides a significant survival benefit vs mitoxantrone + P in pts with mCRPC (Phase III TROPIC study [NCT00417079]; hazard ratio 0.70; p < 0.0001). These findings supported the initiation of ongoing Sanofi-funded CUP and EAP (NCT01254279) to provide access to Cbz prior to commercialization and to collect real-life safety data. Methods: Expected enrollment is ~1600 pts with mCRPC from 250 centers worldwide. Pts receive Cbz (25 mg/m2 Q3W) + P (10 mg oral QD) until progressive disease (PD), death, unacceptable toxicity, physician/pt decision or Cbz commercial availability. Pts are followed until 30 days after last dose. Granulocyte colony-stimulating factor (G-CSF) use is recommended as per ASCO guidance. Results: Interim baseline and safety data from the first 1301 pts treated in 37 countries are now available. Mean age was 68 yrs (22% were ≥ 75 yrs). All pts had an ECOG performance status ≤ 2. Median time from initial prostate cancer diagnosis was 57.6 months and 60% of pts had ≥ 2 metastatic sites; the most common were bone (91%) and lymph nodes (regional 30%, distant 27%). In total, 17% had PD whilst on docetaxel. The median number of Cbz cycles was 6 (range 1–22); median relative dose intensity was 99%. Overall, 837 pts (64%) received G-CSF (n = 123 curative [C], n = 765 prophylactic [P] and n = 99 [C + P]). Of 1142 pts (88%) who discontinued Cbz + P, the most common reasons were PD (44%), adverse event (AE; 27%), physician decision (13%) and commercial availability of Cbz (7%). Grade 3–4 AEs possibly related to Cbz + P occurred in 43% of pts; the most frequent were clinical neutropenia (18%), febrile neutropenia (FN; 7%) and diarrhea (4%). Of 80 pts (6%) with AEs leading to death, the AE was related to Cbz + P in 39 pts (3%). Conclusions: These results provide valuable data on Cbz + P treatment in routine clinical practice, confirming the safety results of clinical trials and showing that treatment with Cbz + P is associated with a manageable safety profile. The incidence of FN seems slightly lower than in TROPIC, owing to more frequent use of G-CSF prophylaxis in the CUP and EAP. Clinical trial information: NCT01254279.

Background Cabazitaxel (CBZ), a novel taxane developed to overcome docetaxel resistance, has been shown to prolong overall survival compared with mitoxantrone in mCRPC patients (pts) who progressed during or after a docetaxel (D)-based therapy. Compassionate Use (CUP) and Early-Access (EAP) Programs are allowing access to the drug before its commercial availability. This interim analysis of the European part of the CUP/EAP focuses on the safety profile of CBZ in elderly population. Methods As of September 1st 2011, 746 pts (range 44-86 years; ≥70 years, n = 325; Results Baseline clinical characteristics of pts aged ≥70 years (group A) or Conclusions This interim analysis of European CUP/EAP suggests that Cabazitaxel-related side effects are manageable in elderly population with mCRPC. de Bono et al. Lancet 2010; 376:1147–54 Disclosure A. Heidenreich: Advisory relationship with Astellas, Jansen Cilag, Sanofi, Ipsen, Takeda Honoraria received from Amgen, Astellas, Glaxo, Ipsen, Jansen Cilag, Pfizer, Sanofi, Takeda, S. Bracarda: Advisory relationship with Sanofi and Janssen and honoraria from Sanofi, M. Mason: Honoraria and lecture fees from Sanofi, Ferring,BMS, Takeda, Janssen Chair and member of advisory boards for Sanofi, L. Sengelov: Investigator in Sanofi clinical trial, W. Gerritsen: Member of advisory board for Sanofi, C. Papandreou: I participated to 2 advisory boards for Sanofi, S. Fossa: Honoraria from Janssen and Sanofi, S. Hitier: Sanofi employee, M. Climent: Advisory relationship and honoraria from Sanofi. All other authors have declared no conflicts of interest.

We surveyed the vegetation and bird faunas of forests on 65 islands in the Gwaii Haanas archipelago of British Columbia, Canada, ranging in size from 1 to > 100000 ha, using point counts at a uniform distance from the shore. Variation in habitat structure was correlated with variation in island area and isolation. Only among the smaller islands did the number of bird species recorded decrease with area. As some species became rarer with decreasing island size, others became more common. The distribution of bird species among the islands was correlated with the distribution of habitat features that were consistent with the biology and ecology of each species. In only a minority of species was their distribution related to area and isolation per se rather than to habitat features correlated with island size and isolation. Hence, we considered that variation in habitat structure mediated by area and isolation was the key factor involved in determining the local composition of the bird community. Only for a few species restricted to the largest islands, or missing from the very small islands, were high rates of extinction related to small population size the most parsimonious hypothesis explaining species distribution patterns. Our results emphasise how considering only the relationship between numbers of species and island area can mask all but the roughest species distribution patterns and prevent a deeper understanding of the biology of islands