Systems Analysis of BCL2 Protein Family Interactions Establishes a Model to Predict Responses to Chemotherapy

Abstract

Apoptotic desensitization is a hallmark of cancer cells, but present knowledge of molecular systems controlling apoptosis has yet to provide significant prognostic insights. Here, we report findings from a systems study of the intrinsic pathway of apoptosis by BCL2 family proteins and clinical translation of its findings into a model with applications in colorectal cancer (CRC). By determining absolute protein quantifications in CRC cells and patient tumor samples, we found that BAK and BAX were expressed more highly than their antiapoptotic inhibitors. This counterintuitive finding suggested that sole inhibition of effector BAX and BAK could not be sufficient for systems stability in nonstressed cells. Assuming a model of direct effector activation by BH3-only proteins, we calculated that the amount of stress-induced BH3-only proteins required to activate mitochondrial apoptosis could predict individual death responses of CRC cells to 5-fluorouracil/oxaliplatin. Applying this model predictor to protein profiles in tumor and matched normal tissue samples from 26 patients with CRCs, we found that differences in protein quantities were sufficient to model the increased tumor sensitivity to chemotherapy compared with normal tissue. In addition, these differences were sufficient to differentiate clinical responders from nonresponders with high confidence. Applications of our model, termed DR_MOMP, were used to assess the impact of apoptosis-sensitizing dugs in lowering the necessary dose of state-of-the-art chemotherapy in individual patients. Together, our findings offer a ready clinical tool with the potential to tailor chemotherapy to individual patients. Cancer Res; 73(2); 519-28. Ó2012 AACR.

Doxorubicin-induced chronic dilated cardiomyopathy—the apoptosis hypothesis revisited

“…Indeed, with such quantitative signal transduction models, we were previously able to predict patient specific susceptibility to chemotherapy [31,44,54,84,85]. It is therefore expected that these models in combination with biomolecular studies that investigate molecular mediators of apoptosis can further help to identify patient more vulnerable to DOX-induced cardiac apoptosis and DCM.…”

“…In the context of cancer cells this mechanism was coined as apoptosis priming, aiming to explain why cancer cells are more susceptible to chemotherapy than normal cells. Specifically, proponents of the priming hypothesis such as the Letai group suggest that cancer cells increase their apoptosis likelihood in response to continuous somatic attack by the immune system [45,54], making them more likely than somatic cells to respond to cytotoxic chemotherapy. Indeed, we observe some parallelism between cancer cells primed by somatic apoptosis [45,55] and cardiomyocyte being exposed to (sub-lethal) DOX with respect to metabolic programming and apoptosis sensitization ( Figure 2).…”

“…In addition, as cancer cells are continuously attacked by the immune system, the constant somatic stress makes them more sensitized to undergo chemotherapy-induced apoptosis. This sensitization was coined as apoptosis priming [45,55] and was recently explained by us through systematic change in balance in favor of pro-apoptotic proteins using a systems analysis [54]. MOMP apoptosis, APOPTO-CELL.…”

On the personalised modelling of cancer signalling * *Supported by EU FP7 grant “SynSignal” (No. 613879).

“…Recently, multiple studies have shown that patient-specific differences in the dynamic behaviour of these signalling networks underlie individual pathogenetic changes and disease manifestation (Fey et al, 2015;Flanagan et al, 2015;Lindner et al, 2013;Murphy et al, 2013). For example, a dynamic model of the JNK pathway could predict the survival probabilities of cancer patients in neuroblastoma, a common childhood cancer (Fey et al, 2015).…”

“…In fact, all the personalised models mentioned are based on this simple principle of directly using the measured mRNA or protein concentrations as static parameters in the model (Fey et al, 2015;Flanagan et al, 2015;Lindner et al, 2013;Murphy et al, 2013). Thus, all these personalised models are based on the assumption that the Supported by EU FP7 grant "SynSignal" (No.…”