Multiple myeloma (MM) is an age-related incurable plasma cell cancer primarily located in the bone marrow. As the second most frequent haematological malignancy it is diagnosed in more than 40 000 patients per year in Western countries (Ferlay et al, 2013; Siegel et al, 2013), and due to aging populations its prevalence is expected to increase (Palumbo et al, 2011; Morgan et al, 2012). Over the past decade, significant advances in treatment options and patient outcome have been achieved with the introduction of proteasome inhibitors and immunomodulatory drugs (Palumbo & Anderson, 2011; Ludwig & Sonneveld, 2012; Kumar et al, 2013; Usmani et al, 2013). However, targeted molecular therapies against oncogenic drivers, which have been developed for the treatment of certain solid tumours and leukaemias have not yet been forthcoming for MM. Indeed, while advanced sequencing projects have now provided unprecedented insights into MM genetics and potential tumour-specific molecular targets, they have also highlighted the impressive molecular heterogeneity underlying this disease (Chapman et al, 2011; Chesi & Bergsagel, 2011; Kuehl & Bergsagel, 2012; Morgan et al, 2012; Leich et al, 2013). A thorough characterization of promising molecular targets for MM subgroups and the development of appropriate inhibitors for beneficial combination treatments are therefore required.
The multi-functional class I phosphatidylinositide 3-kinases (PI3Ks) represent a major hub for the regulation of cell growth and survival, relaying cell surface receptor signals to intracellular signalling cascades (Downward, 2004; Pollak, 2012). Upon activation, PI3Ks catalyse the generation of phosphoinositol lipid substrates which act as second messengers for downstream signalling effector proteins. PI3Ks are classified according to their catalytic isoforms as class IA (PIK3CA, PIK3CB, PIK3CD) and class IB (PIK3CG) (reviewed in Vanhaesebroeck et al, 2010), and their illegitimate activation has been implicated in a variety of cancers, including MM (Vivanco & Sawyers, 2002; Wetzker & Rommel, 2004; Bartholomeusz & Gonzalez-Angulo, 2012). However, genetic lesions that directly affect genes for PI3K isoforms do not occur in MM (Müller et al, 2007; Ismail et al, 2009; Chapman et al, 2011; Leich et al, 2013), although oncogenic signalling via the PI3K/Akt axis is a prominent feature in MM cells (Hsu et al, 2001; Zöllinger et al, 2008; Baumann et al, 2009; Steinbrunn et al, 2011; Ramakrishnan et al, 2012; Munugalavadla et al, 2014). Inhibition of PI3K activity may therefore represent a promising therapeutic strategy to target the disease in a large subgroup of MM patients, and we and others have previously demonstrated that abrogation of PI3K-dependent signalling impairs MM cell survival (Hsu et al, 2001; Pene et al, 2002; Zöllinger et al, 2008; Steinbrunn et al, 2011, 2012; Stengel et al, 2012; De et al, 2013; Azab et al, 2014; Munugalavadla et al, 2014). With the development of novel isoform-specific PI3K inhibitors, more selective targeting of the relevant isoforms has become possible, potentially permitting enhanced anti-tumour activity and a reduction of off-target effects. The aim of our study was therefore to assess the individual contributions of the PI3K isoforms to oncogenic signalling by PI3K in MM and their relevance for the survival of MM cells. We provide evidence that PIK3CA is the major mediator of PI3K-dependent effects in MM and that it should represent the foremost target for pharmacological PI3K isoform blockade. Moreover, combinations of the PIK3CA-specific inhibitor BYL-719 (Furet et al, 2013) with clinically established anti-MM drugs show synergistic modes of action arguing in favor of initiating clinical trials with PIK3CA inhibitors in this disease.