The dual PI3K|[sol]|mTOR inhibitor, NVP-BEZ235, is efficacious against follicular lymphoma

Follicular lymphoma (FL) is one of the most common forms of low-grade non-Hodgkin lymphoma in the Western hemisphere. It usually follows an indolent course. However, 25–60% of cases may transform to aggressive diffuse large B cell lymphomas. FLs are characterized by the t(14;18)(q32;q21) translocation, which results in the overexpression of a chimerical bcl-2/IgH transcript in FL cells. Overexpression of bcl-2 provides a constitutive survival signal for FL. The acquisition of additional mutations in genes, such as p53, leads to FL transformation to diffuse large B cell lymphomas. Although FL is sensitive to some chemotherapeutic agents, relapse is common and transformation to diffuse large B cell lymphomas is considered incurable with conventional chemotherapy. Hence, novel therapeutic agents are urgently needed to improve patient outcomes. The PI3K/Akt/mTOR pathway has a critical role in cell survival and proliferation, and is often activated in many types of cancer. PI3K is activated upon growth factor binding to their cognate receptors. Activated PI3K leads to the activation of PI3K-dependent kinase, which phosphorylates and activates Akt.1 Phosphorylation of Akt at both serine 473 and threonine 308 is required for full activation. Akt activates mTOR by inhibiting tuberous sclerosis complex 2, which heterodimerizes with tuberous sclerosis complex 1, thereby inactivating Rheb and mTOR. mTOR associates with Raptor (mTORC1 complex) to phosphorylate the p70 S6 kinase (S6K), which in turn phosphorylates the S6 ribosomal protein, as well as the eIF4E inhibitor, 4E-BP1, leading to protein translation. In addition, mTOR associates with Rictor (mTORC2 complex) and functions in a feedback loop to phosphorylate and activate Akt at serine 473.1 Akt can also upregulate Bcl-2 expression through cAMP-response element-binding protein. Thus, activation of the PI3K/Akt/mTOR pathway leads to protein synthesis and cell proliferation, as well as cell survival. FL shows phosphorylated and activated Akt and mTOR kinases.2, 3, 4, 5 Downstream targets of mTOR, such as S6K and 4E-BP1, are also phosphorylated. These are sensitive to the single mTOR inhibitor, rapamycin.2, 3, 5 Activation of the PI3K/Akt/mTOR pathway is required for FL growth as rapamycin inhibits the growth of FL cell lines in vitro, and in xenograft models.4 However, rapamycin studies with other cancers have yielded mixed results because rapamycin inhibits mTORC1, but not mTORC2, and recent literature suggests there is increased mTORC2-mediated phosphorylation of Akt in the presence of rapamycin, which may compensate for mTORC1 inhibition by rapamycin.6 As FL shows an activated PI3K/Akt/mTOR pathway,2, 3 we investigated whether NVP-BEZ235, a dual PI3K and mTOR inhibitor, would be effective in inhibiting FL proliferation. NVP-BEZ235 inhibits the kinase activity of PI3K and mTOR by binding, reversibly and competitively, to their ATP-binding clefts.7 We first tested the ability of NVP-BEZ235 to inhibit the growth of the transformed FL cell lines, SUDHL16, FL-18, SUDHL4 and K422. Cells were treated with either 25 or 50 nM NVP-BEZ235 for 3 days and 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt (MTS) readings were taken every 24 h. Each experiment was carried out in triplicate and standard deviation was calculated. Both these drug doses substantially inhibited cell growth of K422, SUDHL16 and FL-18 (Figure 1a). Similar results were obtained with SUDHL4 (data not shown). The downstream targets of PI3K, including Akt, mTOR and S6K, have previously been reported to be activated and phosphorylated in FL.2, 3 We also found that these kinases were phosphorylated in all the transformed FL cell lines: SUDHL4, K422, SUDHL16 and FL-18 (Figure 1b). Dual inhibition of PI3K and mTOR with NVP-BEZ235 for either 16 or 24 h inhibited Akt, mTOR and S6K phosphorylation in all cell lines. Reduced phosphorylation of S6K indicates that NVP-BEZ235 inhibited protein synthesis, whereas reduced phosphorylation of Akt at serine 473 suggests that mTORC2 feedback phosphorylation of Akt was also inhibited. In addition, we carried out an inhibitory constant 50 dose experiment on the FL-18, SUDHL4, K422, and SUDHL16 cell lines. The inhibitory constant 50 of NVP-BEZ235 was 6.59±0.41 nM against FL-18 (Figure 2a), and 6.81±1.03, 5.86±0.57 and 2.4±0.97 nM against SUDHL4, SUDHL16 and K422, respectively, (data not shown). The inhibition of cell proliferation was because of the increased apoptosis, as the NVP-BEZ235-treated cells showed 1.6–2-fold increase of caspase-3 activation after 24 h incubation with the drug, compared with cells treated with vehicle (dimethyl sulfoxide) alone (Figure 2b). Previously, it has been reported that the SUDHL4 cells are resistant to treatment with the proteasome inhibitor, bortezomib.8 We treated SUDHL4 cells with bortezomib (purchased from LC Laboratories, Woburn, MA, USA), either alone or in combination with NVP-BEZ235 (Figure 2c) for 48 h and determined cell survival by MTS assay. We found that bortezomib and NVP-BEZ235 synergistically inhibit the survival of SUDHL4 cells compared with treatment with either drug alone (Figure 2c). The antitumor activity of NVP-BEZ235 was studied in a mouse xenograft model established by subcutaneously injecting 4–5-week-old athymic nude-Foxn1nu female mice with 1 × 106 FL-18 cells. On development of palpable tumors, the animals were split into two groups and treated 5–6 days a week with either vehicle or 45 mg/kg NVP-BEZ235 through oral gavage as previously reported.9 Treatment of FL-18 tumors with NVP-BEZ235 suppressed tumor growth in vivo to a significant degree (Figure 2d) with P0.05, as calculated by linear mixed-effects model. Before the termination of the experiment, the mice in the vehicle-treated group weighed an average of 22.56±2.24 gm per mouse, and the mice in the NVP-BEZ235-treated group weighed an average of 22.50±2.33 gm per mouse (data not shown). Thus, minimal toxicity was seen in the drug-treated mice compared with the control group. The mice were killed after maximal tumor size was reached in the control group. Tumors were excised, sectioned and subjected to immunohistochemical analysis. We found reduced phosphorylated levels of the downstream effectors of PI3K and mTOR, that is, Akt and S6K, respectively, in NVP-BEZ235-treated mice compared with vehicle-control treated mice (Figure 2d). This suggests that the drug was inhibiting its targets effectively. In conclusion, although rapamycin and its derivatives seem to be efficacious against FL,2, 3, 4, 5 they only inhibit mTORC1, but not mTORC2 activity. Given the fact that rapamycin trials in other cancers have lead to the emergence of rapamycin resistance tumors,10 a dual PI3K/mTOR inhibitor like NVP-BEZ235 may prove more effective in the clinic than single mTOR inhibitors. The inhibitory constant 50 for NVP-BEZ235 was much lower than the inhibitory constant 50 for solid tumors, which ranged from 15 to 20 nM.11 Additionally, we found that NVP-BEZ235 and bortezomib synergize to inhibit SUDHL4 proliferation. Hence, our data suggest that NVP-BEZ235 has significant potential for treatment of FL. The authors declare no conflict of interest. This work was supported by NIH grant RO1-CA096500 and UCRF funding to BD, and funding from the Leukemia and Lymphoma society (6021) and the AIDS malignancy consortium to DD. BD is a Leukemia and Lymphoma Society Scholar and a Burroughs Wellcome Fund Investigator in Infectious Disease. We acknowledge Novartis for providing us with NVP-BEZ235. We thank Charlene Ross and the Animal Studies core for their help. We thank Drs Thomas Shea and Kristy Richards as well as members of the Damania and Dittmer labs for helpful discussions.