Our recent Opinion article discussed the oncogenic effects of engaging apoptosis and their impact on cancer (Nat. Rev. Cancer 16, 539–548; 2016)1. We would like to thank Roger Chammas, Luciana Nogueira de Sousa Andrade and Sonia Jancar for their correspondence on our article (Nat. Rev. Cancer (2017) doi:10.1038/nrc.2017.15)2. Caspase protease activity is essential for apoptotic cell death. Among hundreds of diverse substrates, caspases cleave Ca2+-independent phospholipase A2 (iPLA2) leading to its activation3,4; active iPLA2 cleaves phosphatidylcholine to give arachidonic acid and lysophosphatidylcholine (LPC). Through cyclooxygenase enzymatic activity, arachidonic acid can be converted into prostaglandin E2 (PGE2) — a prostanoid that has tumour-promoting properties. Consequently, this provides one means whereby caspase-dependent apoptosis could promote cancer5.

Chammas and colleagues2 highlight oncogenic functions for platelet-activating factor receptor (PAFR) signalling; this is mediated following receptor binding to platelet-activating factor (PAF) and related PAFR ligands. Various cell-killing anticancer therapies, notably radiation, are strong inducers of PAF and PAFR ligands. As the authors discuss, during apoptosis PAF might be produced from LPC (generated by caspase-activated iPLA2) through the action of lysophosphatidylcholine acyltransferases (LPCATs). Although we find this possibility interesting, it is unknown whether apoptotic cells generate PAF in a caspase-dependent manner. Indeed, radiation and chemotherapies induce PAF and PAFR ligands in a non-enzymatic manner (dependent on phospholipid oxidation)6,7, arguing against a major role for caspase-dependent generation of PAF. Given this, the rationale for directly targeting PAFR signalling in cancer therapy seems more compelling than that for inhibiting putative, caspase-dependent PAF generation.