29IN - What is the feature of an optimal new agent: Selective target versus multi-targeted versus combination drugs

Date 01 October 2012
Event ESMO Congress 2012
Session A paradigm shift in early drug development: Individualizing to more patient benefit
Topics Drug Development
Presenter Jan Schellens
Authors J. Schellens, S. Marchetti
  • Department Of Clinical Pharmacology, The Netherlands Cancer Institute, 1066CX - Amsterdam/NL


There are in principle two pharmacological ways to address the question whether selective targeted or multitargeted drugs are to be preferred as single agent and/or in combination. One approach is to focus on benefit and the other is to focus on safety. The modern era of development of targeted agents started with the launch of two key molecules: one monoclonal antibody (MoAb) rituxumab and one so-called small molecule imatinib. Rituximab was registered in the EU in 1998 and since then about seven other MoAbs have been registered. All MoAbs tend to have a selective mechanism of action and target on average one key protein, for example CD-20 (rituximab), or Her2 (trastuzumab). The benefit of these drugs has been clearly demonstrated and the safety of these MoAbs is on average good. Possible exception is where they target the same proteins in normal tissues such as the skin (cetuximab, panitumumab), or the heart (trastuzumab) resulting in sometimes poorly predictable but on average manageable side-effects. The small molecule imatinib was registered in the EU in 2001 and since then more than 13 other targeted small molecules have been registered. Some of these drugs are rather selective, such as erlotinib (mainly EGFR) and vemurafenib (mainly BRAF V600) whereas others are less selective and certainly less selective than the MoAbs, for example sorafenib, pazopanib and sunitinib. This translates into sometimes significant normal tissue toxicity (sunitinib). From the safety perspective one would prefer selective drugs targeting unique tumor expressed proteins or deranged pathways, which would result in a wide therapeutic window. From a benefit perspective at first glance one might prefer multi targeted drugs as tumors show a multitude of mutations and amplifications and a double- or even multi-barreled shotgun approach might hit one or more of these targets at the same time. The disadvantage however is that the mutational profile may not fit to the activity spectrum of the “targeted” drug and that this approach will probably increase normal tissue toxicity. In view of increasing insight in the molecular derangements of tumors one would prefer to have many selective anticancer drugs that can safely be combined on the basis of a patient's individual tumor genetic profile. This would fit into the concept of a “personalized treatment”.


J.H.M. Schellens: Research grants have been received from: Roche, Eisai, GSK and Astrazeneca.

S. Marchetti: No conflicts of interest to declare