Combined mammalian target of rapamycin and vascular endothelial growth factor pathway inhibition in pancreatic neuroendocrine tumors: more than the sum of its parts?

Abstract

Over the last 10 years, major strides have been made in the treatment and understanding of pancreatic neuroendocrine tumors (PNETs). Now considered biologically and genetically distinct from other well-differentiated neuroendocrine tumors (NETs), the therapeutic landscape for unresectable PNETs has evolved rapidly over the past several years. Sunitinib and everolimus were approved for use in advanced disease in 2011 on the basis of a 6-month improvement in progression-free survival (PFS). It is noteworthy that both agents are cytostatic and that their capacity to induce tumor shrinkage is low (objective response rate [RR], 10%). Recent phase III data suggest that lanreotide (a somatostatin analog) also delays progression in nonfunctional well-differentiated NETs (including those of pancreatic origin). In terms of strategies that can shrink rather than just stabilize tumors, liver-directed therapy is often used in patients with liver-dominant disease. Prospective randomized trials are lacking, but the data suggest that embolization (bland, chemoembolization, or radioembolization) is associated with a roughly 50% RR. Chemotherapy is often used in a subset of patients (typically with bulky, refractory, and/or relatively rapidly growing tumors), but the optimal regimen is unclear. Streptozocin-based treatment was approved for PNETs in the 1990s, but the merits of this approach have been called into question by subsequent trials. In the last few years, temozolomide-based treatment has gained traction in patients with advanced PNETs, despite the lack of data from randomized controlled trials. Radiographic RRs of up to 70% have been reported in retrospective trials, but the data from prospective trials have been less impressive (RRs in the 24% to 45% range). The Eastern Cooperative Oncology Group (ECOG)/American College of Radiology Imaging Network (ACRIN) trial E2211 (a randomized phase II study of temozolomide or temozolomide and capecitabine in patients with advanced PNETS) is an ongoing randomized phase II trial designed to assess the value of temozolomide-based therapy in patients with progressive PNETs. Given the relatively modest activity of single-agent everolimus and sunitinib in PNETs, there is interest in developing novel treatment strategies, including mammalian target of rapamycin (mTOR) and vascular endothelial growth factor (VEGF) pathway inhibitor combinations. Concurrent therapies might prevent or delay resistance by blocking multiple signaling pathways. The PI3K/AKT/mTOR pathway is frequently activated in cancer and plays a critical role in cell growth, metabolism, and survival. The signaling pathway is extremely complex because it is regulated by several different feedback loops; furthermore, PI3K/AKT/mTOR signaling interacts with other signaling cascades. Rapalogs such as temsirolimus and everolimus inhibit activation of TORC1, a key regulator of cell growth, proliferation, survival, protein synthesis, and transcription. They are thought to inhibit VEGF signaling indirectly, in contrast to bevacizumab and VEGF receptor tyrosine kinase inhibitors (TKIs). Combination therapy might lead to additive or synergistic effects by producing more complete inhibition of a single pathway (eg, VEGF signaling) or by simultaneously inhibiting two or more distinct pathways, thus yielding more robust antitumor effects and even tumor shrinkage. In the article that accompanies this editorial, Hobday et al report results from a multicenter trial of temsirolimus once per week and bevacizumab every other week in 56 patients with progressive PNETs. Importantly, patients were eligible for inclusion only if they had evidence for progressive disease by RECIST criteria within 7 months of study entry, suggesting enrichment for relatively aggressive disease (compared with the pivotal trials of sunitinib and everolimus). In the phase III study of sunitinib, progression by RECIST criteria within the preceding 12 months was required; in the everolimus study, any radiographic progression within 12 months was sufficient. Prior treatment had to be completed 4 or more weeks before registration. There was no mention of a requirement for measurable disease outside the treated area in patients with prior liver-directed therapy or restrictions related to prior treatment with peptide receptor radiotherapy. Given the potential for delayed responses, it would be useful to know whether the six patients who underwent these treatments (four, chemoembolization; two, peptide receptor radiotherapy) did so immediately before enrolling onto this study and whether they were counted as responders. Finally, 29 patients had previously been treated with octreotide, but concurrent octreotide during protocol therapy was not reported (a potentially important factor if concomitant octreotide either hinders or augments therapeutic efficacy). JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 33 NUMBER 14 MAY 1