New Oncological Drugs: Analysis of Survival Gain

Abstract

lon retto (9,7%). Il cancro allo stomaco viene al quarto posto (7,8%); da notare che, nelle prime stime del 1975, esso rappresentava invece la neoplasia più diffusa. Segue il cancro alla prostata (7,2%), con circa tre quarti dei casi registrati nei Paesi a sviluppo avanzato.Passando dall'incidenza alla mortalità, si stimano 7,6 milioni di decessi. Anche qui, il cancro ai polmoni è al primo posto come fattore causante (18,2%). Seguono poi il cancro allo stomaco (9,7%), quello al fegato (9,2%), quello al colon retto (8%) e quello al seno (6,1%) (1).Tutte queste stime non sono del tutto confrontabili con quelle pubblicate in periodi precedenti, a motivo della crescente disponibilità di dati e dei miglioramenti nei metodi di stima. Con questa premessa, è comunque indicativo il confronto tra il dato globale d'incidenza sopra riportato (12,7 milioni di casi, su un insieme di 27 principali tipologie di cancro considerate) e l'analogo stimato per il 1980: 6,4 milioni di casi (su 16 tipologie) (2). L'incidenza della malattia appare dunque in forte crescita. Tenendo conto di due processi: 1) l'aumento della popolazione mondiale (combinato allo spostamento new oncological drugs: analysis of survival gain Cancer incidence is growing worldwide, boosting substantial investment in research and development for ever better therapies. The present study aims to evaluate the benefits from innovative drugs for oncology in terms of overall survival (OS, a gold standard), updating former publication findings on this issue. Survival gain was defined as the difference in median OS between research and control arms, respectively. Starting from regulatory approvals for reimbursement issued by AIFA (Agenzia Italiana del Farmaco) and officially published between January 2010 and July 2013, 39 pivotal trials were identified on antineoplastic agents (21 related to new chemical entities and 18 to therapeutic extensions). In seven studies, OS data was not reported nor could be reconstructed; so they were not included in the analysis. Among the remaining 32 studies, the survival gain was statistically significant (p <0.05) in 12 studies. It ranged from 22 months (rituximab + fludarabine + cyclophosphamide vs fludarabine + cyclophosphamide in patients with chronic lymphocytic leukaemia) to 2.2 months (nab-paclitaxel vs conventional paclitaxel in women with breast cancer); the median value of the gain was 3.2 months and the mean value was 5.9 months (95% CI: 2.5-9.3). In the 20 studies where significance was not found in the median OS difference (though in most cases it was found if time-dependent endpoints were used as outcome measures), the survival gain ranged from 106 months (nilotinib vs imatinib in patients with chronic myeloid leukemia) to -2 (i.e. a survival loss) months (vandenatib vs placebo in patients with locally advanced or metastatic medullary thyroid cancer); the median value was 2.3 months and the mean value was 14.5 months (95% CI:1.2-27.8). Though the median survival gains in both groups may look modest in absolute terms, they ma...