In the past two decades, the number of targeted therapies for cancer treatment has grown exponentially. SMAC (second mitochondria-derived activator of caspases) mimetics comprise a class of targeted therapies that sensitize tumor cells to apoptosis [1], [2]. They mimic an endogenous molecule, SMAC/Diablo, produced by mitochondria that antagonizes IAPs (inhibitor of apoptosis proteins), thus halting apoptosis downstream of intrinsic and extrinsic factors [3]. In non-cancerous cells, SMAC is released from the mitochondria and binds to IAPs, thereby allowing the cell to complete apoptosis via caspases. However, some cancers upregulate IAPs as a tactic to evade apoptosis signals. XIAP can directly antagonize caspase enzymatic activity, thereby overcoming terminal components in apoptotic signaling. The anti-apoptotic function of cIAPs is thought to be more indirect. cIAPs’ RING finger domain endows them with ubiquitin ligase activity, which affects signaling downstream of TNF receptors. IAPs simultaneously activate canonical NF-kB signaling and inhibit activation of the death-inducing complex II [4]. Thus, inhibition of IAPs with SMAC mimetics sensitizes cancer cells to intrinsic and extrinsic apoptotic signals.

The ability of SMAC mimetics to bind IAPs and promote cancer cell apoptosis in vitro prompted interest in their clinical application [5], [6], [7], [8], [9], [10]. In preclinical in vivo animal models, SMAC mimetics showed encouraging results, particularly when used in tandem with chemotherapy drugs to sensitize cancer cells to chemotherapy [6], [8], [9], [10], [11], [12], [13]. However, in clinical trials, SMAC mimetics have yet to show efficacy as a monotherapy or to increase efficacy in combination with chemotherapy or immunotherapy (Supplemental Table 1) [14].

In addition to inhibiting apoptosis, IAPs also play a role in inhibiting non-canonical NF-kB activation in immune cells [15], [16]. Thus, SMAC mimetics, which inhibit IAPs, have been predicted to increase non-canonical NF-kB activation [17], [18], [19], [20], [21]. This is particularly relevant to cancer immunotherapy, because non-canonical NF-kB has been shown to be necessary for optimal T cell responses. In mouse T cells, activation of non-canonical NF-kB downstream of TNF receptor family members promotes T cell memory and effector functions [22], [23], [24], [25], [26], [27]. SMAC mimetics do, indeed, activate non-canonical NF-kB in immune cells [17], [28], [29], and this was associated with enhanced T cell responses in mice in vivo and in humans in vitro [17], [28]. In addition, SMAC mimetics have been shown to synergize with immunotherapy in mouse models of cancer [17], [18], [30], [31], and this effect depended on the presence of T cells [18], [31]. These studies suggest that SMAC mimetics might have two different anti-cancer mechanisms—a direct effect on cancer cell viability and an indirect effect via activating anti-tumor T cell responses. If true, this would increase enthusiasm for SMAC mimetics as targeted therapies and could have implications for combining SMAC mimetics with immunotherapy.

The aim of the present study was to directly compare the effects of three different SMAC mimetics on multiple measures of human CD4 and CD8 T cell function. To our surprise, we found that all three inhibited T cell proliferation under several culture conditions. Moreover, contrary to previous studies, none of the SMAC mimetics increased secreted or intracellular production of pro-inflammatory cytokines. Instead, birinapant, the most common SMAC mimetic in clinical trials, decreased percentages of polyfunctional T cells that secrete two or more cytokines. These data suggest that, in humans, SMAC mimetics are unlikely to promote anti-tumor T cell responses, and thus may not synergize with immunotherapy as has been proposed.