¹⁸F-FDG Is a Surrogate Marker of Therapy Response and Tumor Recovery after Drug Withdrawal during Treatment with a Dual PI3K/mTOR Inhibitor in a Preclinical Model of Cisplatin-Resistant Ovarian Cancer^,

Abstract

AIM: Targeting the phosphoinositide 3-kinase (PI3K)/mammalian target of rapamycin (mTOR) pathway is a potential means of overcoming chemoresistance in ovarian cancer. We investigated the capability of ¹⁸F-fluororodeoxyglucose (¹⁸F-FDG) small-animal positron emission tomography (SA-PET) to predict the effects of a dual PI3K/mTOR inhibitor (BEZ-235) in a cisplatin-resistant ovarian cancer model. METHODS: In a first experiment, nude rats bearing subcutaneous SKOV3 tumors received BEZ-235 for 3 days given alone or after paclitaxel and were compared to controls (either untreated or that were given the excipients of paclitaxel and BEZ-235). SA-PET was performed at baseline, on day 3, and day 7. In a second experiment aiming at further exploring the kinetics of ¹⁸F-FDG tumor uptake during the first 48 hours following drug cessation, untreated controls were compared to rats receiving BEZ-235, which were imaged at baseline, on day 3, on day 4, and on day 5. SA-PET results were compared to cell proliferation assessment (Ki-67), PI3K/mTOR downstream target expression studies (pAKT and phospho-eukaryotic translation initiation factor 4E-binding protein 1), and apoptosis evaluation (cleaved caspase-3). RESULTS: In the first experiment, BEZ-235, compared to untreated controls, induced a marked decrease in ¹⁸F-FDG uptake on day 3, which was correlated to a significant decrease in cell proliferation and to a significant PI3K/mTOR pathway inhibition. No tumor necrosis or apoptosis occurred. Four days following treatment cessation, tumor recovery (in terms of PI3K/mTOR inhibition and cell proliferation) occurred and was identified by ¹⁸F-FDG SA-PET. Paclitaxel plus BEZ-235 showed results similar to BEZ-235 alone. In the second experiment, PI3K/mTOR pathways exhibited partial recovery as early as 24 hours following treatment cessation, but both ¹⁸F-FDG SA-PET and cell proliferation remained unchanged. CONCLUSIONS: ¹⁸F-FDG SA-PET is a surrogate marker of target inhibition during treatment with BEZ-235 and predicts tumor recovery 4 days after drug withdrawal, but not during the first 48 hours following drug cessation, when a lag between PI3K/mTOR pathway recovery and metabolic recovery is observed. ¹⁸F-FDG SA-PET could be used for therapy monitoring of PI3K/mTOR inhibitors, but our results also raise questions regarding the potential impact of the delay between PET imaging and the last drug intake on the accuracy of FDG imaging.

Introduction

Ovarian cancers are the leading cause of death from gynecological malignancies worldwide and the fifth most common cause of cancer death in women [1]. The standard treatment is aggressive surgery followed by platinum/taxane-based chemotherapy. After initial treatment, platinum-resistant cancer recurs in approximately 25% of patients within 6 months [2], and despite an initial response rate of about 70%, most patients relapse and develop resistance to platinum based chemotherapy. This chemoresistance is the main cause of treatment failure and leads to an overall survival at 5 years of less than 30% [3]. Although several alternate options are available for the treatment of platinum-resistant ovarian cancer, including topotecan, liposomal doxorubicin, weekly paclitaxel, and gemcitabine as monotherapy [4], phase III studies suggest a similar moderate efficacy of these different drugs, with response rates estimated to be approximately 20% [5,6]. Therefore, there is a need for new therapeutic strategies, including molecularly targeted therapies, to overcome this chemoresistance.

Recent data showed the biologic rationale and clinical activity of mammalian target of rapamycin (mTOR) inhibitors in gynecological cancers [7]. Indeed, a recent prospective large-scale genomic analysis has shown that the phosphoinositide 3-kinase (PI3K)/AKT pathway is frequently deregulated in high-grade serous ovarian tumors [8]. Furthermore, simultaneous AKT and mTOR activation can be present in up to 87% of ovarian tumors [9]. PI3K can be considered as a major mediator of survival signals that protect ovarian cancer from apoptosis induction [10].

BEZ-235 is a novel oral therapeutic agent that inhibits two key proteins of the PI3K/AKT/mTOR pathway, the PI3K and mTOR proteins, and it is currently evaluated in phase I clinical trials [11]. While no clinical trials using BEZ-235 in the setting of cis-platinum-resistant ovarian tumors are currently registered, preclinical data showed that BEZ-235 decreases cell proliferation in ovarian cancer cell lines and sensitizes cisplatin-resistant cells to the cytotoxic effects of cisplatin [12]. In preclinical studies where it was used as a single agent, this drug seems primarily to promote tumor stasis and delay progression rather than activate apoptotic mechanisms inducing cell death and tumor shrinkage [10]. In in vitro and in vivo studies that combined a PI3K inhibitor (LY294002) with paclitaxel—a drug widely used in ovarian cancer—in ovarian cancer, an increased efficacy of the chemotherapy on tumor cell growth and dissemination compared with either agent alone was observed [13]. Therefore, BEZ-235 given alone or in association with paclitaxel holds promise for the treatment of cis-platinum-resistant ovarian cancer. However, there is no biomarker capable of predicting the benefit of BEZ-235. Indeed, it has been shown that PI3K mutations do not predict for the sensitivity of ovarian cell carcinoma cells to PI3K/AKT/mTOR inhibitors [14]. Positron emission tomography (PET) imaging has been shown to be particularly useful for evaluating the efficacy of molecularly targeted therapies given either alone or in combination with conventional chemotherapies in preclinical studies [15]. In particular, ¹⁸F-fluororodeoxyglucose (¹⁸F-FDG) uptake has been shown to be an early surrogate marker of drug efficacy during mTOR inhibition [16].

The aim of this study was to evaluate the ability of ¹⁸F-FDG PET to predict early response to BEZ-235 given alone or associated to paclitaxel in a rat model of subcutaneously transplanted human cisplatin resistant ovarian cancer. ¹⁸F-FDG PET results were correlated not only to tumor response early after treatment but also to tumor recovery after drug discontinuation.

Materials and Methods

Cell Line Culture

The SKOV3 cell line (American Type Cell Collection, Manassas, VA) was cultured in 450-cm² flasks in RPMI 1640 medium supplemented with 2 mM Glutamax, 25 mM Hepes, 10% fetal calf serum, and 33 mM sodium bicarbonate (Fisher Scientific Bioblock, Illkirch, France) at 37°C in a 95% humidified air, 5% CO₂ atmosphere. All cell culture media and components were from Gibco Life Technologies (Cergy Pontoise, France).

Animal Model

Four-week-old female nude rats (n = 28; Harlan Laboratories, Indianapolis, IN) were injected subcutaneously with the SKOV3 human ovarian cancer cell line. Each animal received four injections: two in the shoulders and two in the upper part of the thighs. We grafted four tumors per animal to limit the number of animals killed and to maximize the number of tumors imaged within a small-animal PET (SA-PET)/computed tomography (CT) imaging session. Syringes were prepared as follows: 10 x 10⁶ SKOV3 cells in 0.2 ml of 50% RPMI/Matrigel (BD Biosciences, Franklin Lakes, NJ). The SKOV3 cell line was chosen because it is a cisplatin-resistant cell line that matches well with the dilemma encountered in the clinical setting. To improve the tumor take rate, animals received a single, 2-Gy dose of whole-body irradiation 24 hours before cell implantation. Animals were kept under pathogen-free conditions and allowed to feed freely. Treatments began when the largest tumor measured 7 to 10 mm in diameter. Animal experiments were approved by the regional Ethics Committee (No. N/02-10-09/18/10-12).

Drug Preparation

Paclitaxel (Mylan Pharma, Canonsburg, PA) was diluted in a peritoneal dialysis solution (Dianeal; Baxter, Deerfield, IL) according to previous studies, and 10 ml of this preparation was administered intraperitoneally at 5mg/kg on day 0. BEZ-235 (Selleck Chemicals, Houston, TX) was diluted with a polyethylene glycol (Sigma-Aldrich, St Louis, MO) and N-methyl-2-pyrrolidone (Sigma-Aldrich) solution and was administered orally at 50 mg/kg on day 1 (24 hours after paclitaxel injection), day 2, and day 3. The control animals received a Cremophor EL (Sigma-Alrich) and ethanol solution on day 0 and an N-methyl-2-pyrrolidone/polyethylene glycol solution from day 1 to day 3, which are the excipients of paclitaxel and BEZ-235, respectively [11].

Study Design

In a first experiment, two cohorts of rats were used (Table 1). The first consisted of 14 rats used for PET imaging on day 0, day 3, and day 7 and molecular analysis on day 7. The second consisted of 14 rats for the correlative experiment to access molecular analysis on day 3. Moreover, one rat was sacrificed on day 0 as a baseline reference and the others on day 3. Both cohorts consisted of five groups: untreated control, control (excipients), BEZ-235, paclitaxel, and paclitaxel plus BEZ-235.

Table 1.

Experimental Design: Number of Animals and Tumor Samples Used for the SA-PET Imaging and Correlative Experiments.

Experiment	Cohort		Day 0	Day 3	Day 4	Day 5	Day 7
Experiment 1	SA-PET/CT	Control (untreated)	2 (8)	2 (8)	-	-	2 (8)
		Control (excipients)	3 (12)	3 (12)	-	-	3 (12)
		Paclitaxel	3 (12)	3 (12)	-	-	3 (12)
		BEZ-235	3 (12)	3 (12)	-	-	1 (4)*
		BEZ-235 + paclitaxel	3 (12)	3 (12)	-	-	2 (8)†
	Correlative studies	Control (untreated)	3 (12)	3 (12)	-	-	‡
		Control (excipients)	3 (12)	3 (12)	-	-	‡
		Paclitaxel	3 (12)	3 (12)	-	-	‡
		BEZ-235	3 (12)	3 (12)	-	-	‡
		BEZ-235 + paclitaxel	3 (12)	3 (12)	-	-	‡
Experiment 2	SA-PET/CT	Control (untreated)	4 (16)	3 (12)	2 (8)	1 (4)	-
		BEZ-235	6 (24)	6 (24)	5 (20)	4 (16)	-
	Correlative studies	Control (untreated)	1 (4)	1 (4)	1 (4)	1 (4)	-
		BEZ-235		1 (4)	1 (4)	4 (16)	-

^*Two animals had to be sacrificed because of significant weight loss.

^†One animal had to be sacrificed because of significant weight loss.

^‡Animals were sacrificed immediately after the SA-PET/CT had been performed and tumors used for correlative studies on day 7.

In a second experiment aiming at further exploring the kinetics of ¹⁸F-FDG tumor uptake following drug cessation, we used a single cohort of 10 animals consisting of untreated controls and rats receiving BEZ-235. In that cohort, animals were used for both PET imaging and molecular analysis. For that purpose, one rat was sacrificed on day 0 as a baseline reference, one animal was sacrificed on day 3 and day 4 in both groups, and all remaining animals were sacrificed after the last PET examination had been performed.

SA-PET/CT Acquisitions and Reconstruction Settings

Animals were kept fasting for 6 hours and were injected intravenously through the tail vein with a 29-gauge needle. Animals received an average activity of 38 ± 7 MBq (first experiment) or 39 ± 5 MBq (second experiment) and were imaged 91 ± 8 minutes (first experiment) or 105 ± 14 minutes (second experiment) following injection.

SA-PET/CT examinations were performed on an Inveon SA-PET/CT (Siemens Medical Solutions, Knoxville, TN). First, SA-CT images were acquired over approximately 10minutes using 80 keV and 500 µA. PET acquisitions were then performed using energy and coincidence timing windows of 350 to 650 keV and 3.4 ns, respectively. Emission scan duration was 10 minutes.

Reconstructions were performed using three-dimensional maximum a posteriori (MAP) reconstruction with a 128 x 128 transaxial image matrix size. Data were corrected for attenuation and scatter events. OSEM-3D/MAP was employed with two OSEM-3D iterations and 18 MAP iterations with the smoothing parameter β set to 0.2.

SA-PET/CT Analysis

Data analysis was performed on a Siemens workstation dedicated to SA-PET/CT interpretation. SA-PET and SA-CT images were interpreted by a researcher who was unaware of the treatments received by the animals.

Three-dimensional regions or volumes of interest (VOIs) were drawn over subcutaneous tumors by means of an isocontour set to 50%. The mean voxel values were extracted from each VOI, and mean standardized uptake values (SUVs) were computed as follows:

$$\text{SUV}=\frac{\text{tumor activity(Bq/ml)}x\text{body weight(g)}}{\text{injected dose(Bq)}}.$$

This kind of VOI is warranted when one wants to take into account tumor heterogeneity. Residual activity of the syringe immediately after injection and activity in the tail in the case that extravasation occurred during tracer injection were subtracted from the activity of the syringe. Activity in the tail was obtained by a VOI encompassing the entire hot spot and corrected for decay to evaluate the activity at the time of injection, assuming that no tracer absorption occurred in the tail during the tracer uptake period.

Tumor volume was evaluated on CT slices according to the following formula:

$${\text{Volume(mm}}^{\text{3}})=\frac{\pi xa(\text{mm})xb(\text{mm})xc(\text{mm})}{\text{6}},$$

where a, b, and c are the long axis on coronal images and two perpendicular small axes at the level where the tumor appeared the largest on transaxial images, respectively.

Western Blot

Cells were rinsed with ice-cold phosphate-buffered saline and lysed in a lysis buffer [50 mM Tris-HCl (pH 8.1), 150 mM NaCl, 1% NP-40, 5 mM EDTA, 10 mM NaF, 4 mM PMSF, 2 mM aprotinin, 10 mM NaPPi, 1 mM Na₃VO₄, and a complete mixture of protease inhibitors (Roche Applied Science, Meylan, France)] and incubated on ice for 30 minutes. Lysates were then collected. After centrifugation (13,200g, 10 minutes, 4°C), protein concentrations were determined using the Bradford assay (Bio-Rad, Hercules, CA). An equivalent total amount of proteins was separated on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (4–15%; Bio-Rad), transferred onto a polyvinylidene fluoride membrane and blocked in 5% BSA. The primary antibodies were incubated overnight at 4°C: phospho-AKT (1:1000), AKT (1:750), phospho-eukaryotic translation initiation factor 4E-binding protein 1 (p4E-BP1; 1:500), 4E-BP1 (1:750) from Cell Signaling Technology (Danvers, MA), actin (Millipore, Billerica, MA; 1:10,000), and tubulin (Sigma; 1:4000). After washing, the membrane was incubated with secondary antibodies conjugated with HRP [anti-rabbit at 1:10000 (Cell Signaling Technology) and anti-mouse at 1:5000 (Amersham, Waukesha, WI)] for 1 hour at room temperature. The bands were visualized with the ECL Prime (Amersham) and the SuperSignal West Femto Chemiluminescent Substrate (Pierce, Rockford, IL).

Immunohistochemistry Studies

Automated immunohistochemistry using a DakoCytomation Autostainer was performed on 4-µm-thick paraffin sections. The polyclonal antibody anti-Ki-67 was obtained from Novocastra (A. Menarini Diagnostics, Antony, France); the polyclonal antibody cleaved caspase-3 (Asp¹⁷⁵) and the monoclonal p4E-BP1 (Thr^37/46) were obtained from Cell Signaling Technology (Ozyme, Saint Quentin Yvelines, France).

Briefly, deparaffinized slides were treated for 30 minutes by the high-temperature heating antigen retrieval technique in 0.01 M citrate buffer (pH 6) for Ki-67 antibody or in CC1 buffer (pH 8; Ventana, Roche Diagnostics, Meylan, France) for cleaved caspase-3 and p4E-BP1 antibodies to unmask epitopes. Sections were incubated with Ki-67 antibody (1:500), cleaved caspase-3 antibody (1:200), or p4E-BP1 antibody (1:50) for 1 hour at room temperature. After washing, slides were incubated with Rabbit IgG Vectastain ABC Kit (Vector Laboratories, Clinisciences, Nanterre, France) according to the manufacturer's instructions. Staining was revealed with DAB chromogen (Vector Laboratories), and sections were counterstained with Hematoxylin QS (Vector Laboratories).

The percentage of stained cells on the whole tumor sections was estimated by an experienced pathologist, who was unaware of the treatment groups.

Statistical Analysis

Data are presented as means ± SD. The change in SUV_mean over the different time points is reported as a percentage of baseline, according to the following formula:

$$\text{Change in SUV}=\frac{\text{SUV on day }x-\text{SUV on baseline}}{\text{SUV on baseline}}x100.$$

Statistical analysis was performed on a per-lesion basis.

In the first experiment, the nonparametric Kruskal-Wallis test was used to compare percentage change in tumor uptake and tumor volume in the control (untreated and excipients) and treated groups (BEZ-235, paclitaxel, and BEZ-235 plus paclitaxel) at specific time points. A post hoc test was performed with the Dunn test for multiple comparisons.

In the second experiment, the nonparametric Mann-Whitney test was used to compare percentage change in tumor uptake and tumor volume in the untreated control and BEZ-235 groups at specific time points.

For all tests, a two-tailed P value of .05 or less was considered statistically significant. Statistical analysis, graphs, and plots were performed with Prism (GraphPad Software, La Jolla, CA).

Results

Impact of Treatments on Tumor Growth

On baseline, tumor volume was lower in the paclitaxel group compared to the BEZ-235 and the paclitaxel plus BEZ-235 groups.

On day 3 and day 7, there was no significant difference in tumor volume, as assessed by CT among the BEZ-235 and the paclitaxel plus BEZ-235 groups and the untreated control group (Figure 1).

Figure 1.

Impact of BEZ-235 on tumor growth. Mean ± SD value in each group is shown.

¹⁸F-FDG Uptake after Treatment by BEZ-235, Paclitaxel, and Paclitaxel Plus BEZ-235

¹⁸F-FDG uptake in tumors of animals that did not receive any treatment (untreated control) continuously increased over time (+37% on day 3 and +59% on day 7; Figure 3).

Figure 3.

BEZ-235 inhibits the PI3K/mTOR pathways. (A) Immunohistochemistry studies of representative tumors. The phosphorylation of 4E-BP1 is markedly reduced on day 3 for the BEZ-235 conditions within all the tumor sections. Staining is similar to the control groups on day 7. (B) Western blot of representative tumors: The administration of BEZ-235 inhibits the phosphorylation of AKT and 4E-BP1 on day 3, and this effect is no longer observed on day 7.

Treatment with either paclitaxel or its excipient induced a small decrease (-19%) in ¹⁸F-FDG tumor uptake on day 3 followed by a return to baseline values on day 7. No difference in percentage change between the control (excipients) and the paclitaxel groups was observed on day 3 or on day 7.

Treatment with BEZ-235 given alone or after paclitaxel induced a marked decrease in ¹⁸F-FDG uptake with a percentage decrease of 68% and 73%, respectively. Noticeably, this marked decrease was homogeneous, as demonstrated by the narrow confidence intervals observed (Figure 2) that can be compared to the larger confidence intervals observed in the other treatment or control groups. This finding is also visible when analyzing data on a per-animal basis (Figure W1). Four days after treatment cessation, a partial tumor recovery occurred, with percentage change of -31% and -26% compared to baseline in the BEZ-235 and paclitaxel plus BEZ-235 groups, respectively. The larger confidence interval observed in the paclitaxel plus BEZ-235 group is related to three of eight tumors (38%) that achieved an almost complete recovery. Visually, treatment by BEZ-235 induced mostly a marked and homogeneous decrease in ¹⁸F-FDG uptake, but a dramatic decrease was also seen at the center of some tumors that was associated with a persistent rim of significant uptake at the tumor edge. In the latter case, the metabolic recovery observed at day 7 clearly began from the edge of the tumor toward its center, as shown in Figure 2B.

Figure 2.

¹⁸F-FDG SA-PET response to BEZ-235 and paclitaxel plus BEZ-235 compared to control groups. (A) Percentage variation (mean ± SD) in SUV_mean compared to baseline is shown on day 3 and day 7. The differences among different groups were tested with the Kruskal-Wallis test, and post hoc test was performed with the Dunn test for multiple comparisons; *, **, and *** indicate two-tailed P < .05, P < .01, and P < .001, respectively. (B) Representative transverse slices at the level of the subcutaneous pelvic or thoracic tumors are shown for the different treatment and control groups.

Effect of BEZ-235 on the PI3K/mTOR Pathways

As shown by Western blot analysis (Figure 3B), BEZ-235 downregulated the key protein pAKT, a downstream marker of PI3K activation. BEZ-235 also effectively decreased phosphorylation of 4E-BP1, a downstream marker of mTOR activation, as shown by immunohistochemistry studies that revealed a marked and homogeneous decrease in the phosphorylation of 4E-BP1 on day 3 on the whole tumor sections (Figure 3A).

On day 7, an almost complete recovery of PI3K/mTOR downstream occurred.

Effect of BEZ-235 on Cell Proliferation and Cell Death

On day 3, treatment with BEZ-235 given alone or after paclitaxel induced a marked decrease in cell proliferation, as assessed by Ki-67 immunostaining, when compared to untreated controls (10% and 8% vs 34%, respectively; P < .01). Compared to all other groups (Figure 4), this decrease in cell proliferation in the BEZ-235 and paclitaxel plus BEZ-235 groups was statistically significant. On day 7, no difference in Ki-67 staining was observed between the control and treated groups and Ki-67 staining almost returned to baseline values.

Figure 4.

BEZ-235 inhibits tumor proliferation. (A) Quantification of Ki-67 staining (percentage of positive cells). The differences among different groups were tested with the Kruskal-Wallis test, and post hoc test was performed with the Dunn test for multiple comparisons; *, **, and *** indicate two-tailed P < .05, P < .01, and P < .001, respectively. (B) Immunohistochemistry studies of representative tumors. Cell proliferation, as assessed by Ki-67 immunostaining, is markedly reduced on day 3 in the BEZ-235 and paclitaxel plus BEZ-235 groups but is similar to control groups on day 7.

To fully characterize cell death, necrosis was studied at pathology and apoptosis by immunohistochemistry (IHC) detection of cleaved caspase-3, an early marker of the caspase cascade. There was no necrosis or apoptosis detected in the treated groups compared to control groups (Figure 5).

Figure 5.

No evidence of cell death (apoptosis, necrosis) is observed after 3 days of treatment with BEZ-235. Immunohistochemistry and hematoxylin eosin safran (HES) studies of representative tumors are shown. No necrosis or cleaved caspase-3 is observed in any of the treatment conditions.

Further Analysis on ¹⁸F-FDG, PI3K/mTOR Pathways, and Cell Proliferation Kinetics following BEZ-235 Withdrawal

¹⁸F-FDG uptake in tumors of animals that did not receive any treatment (untreated control) remained higher than baseline values (+40% on day 3, +19% on day 4, and +19% on day 5). Treatment with BEZ-235 induced a marked decrease in ¹⁸F-FDG uptake with a percentage decrease of -50% on day 3. Similarly to the first experiment, this decrease was homogeneous, as demonstrated by the narrow confidence intervals observed that can be seen when analyzing data on a per-lesion (Figure 6) or on a per-animal basis (Figure W1). The decrease in ¹⁸F-FDG tumor uptake persisted on day 4 (-51%) and day 5 (-46%), and the difference in percentage change in ¹⁸F-FDG in the BEZ-235 and untreated control groups was statistically significant at all time points.

Figure 6.

Further analysis on ¹⁸F-FDG, PI3K/mTOR pathways, and cell proliferation kinetics following BEZ-235 withdrawal. (A) Percentage variation (mean ± SD) in SUV_mean compared to baseline is shown on day 3, day 4, and day 5. The differences among different groups were tested with the Mann-Whitney test. (B) Immunohistochemistry studies of representative tumors. Cell proliferation, as assessed by Ki-67 immunostaining, is markedly reduced from day 3 to day 5 in the BEZ-235 group compared to the control group. In contrast, p4E-BP1 is markedly decreased on day 3 in the BEZ-235 group compared to the control group, but staining is almost similar to the control groups on days 4 and 5. (C) Quantification of Ki-67 staining (percentage of positive cells). The differences among different groups were tested with the Mann-Whitney test. (D) Western blot of representative tumors: The administration of BEZ-235 inhibits the phosphorylation of AKT and 4E-BP1 on day 3. pAKT increased in a time-dependent manner on days 4 and 5, in contrast with p4E-BP1, whose phosphorylation is more clearly visible on day 5.

In contrast with ¹⁸F-FDG PET results, the PI3K/mTOR pathways exhibited partial recovery on day 4 and on day 5, as demonstrated by IHC and Western blot studies demonstrating a trend toward a normalization of the phosphorylation of AKT and 4E-BP1. pAKT increased in a time-dependent manner on days 4 and 5, in contrast with p4E-BP1, whose phosphorylation was more clearly visible on day 5.

From day 3 to day 5, cell proliferation assessment was congruent with ¹⁸F-FDG PET imaging, with a Ki-67 staining significantly lower in the BEZ-235 group than in the untreated control group.

Discussion

As most patients with ovarian cancer relapse within the first few years following platinum-based chemotherapy because of chemotherapy resistance [17], new targeted therapies are urgently needed [18]. Targeting the PI3K/mTOR pathway is a potential means of overcoming chemoresistance in ovarian cancer. The present study shows that ¹⁸F-FDG SA-PET is a surrogate marker of the early response to a dual PI3K/mTOR inhibitor in a preclinical model of cisplatin-resistant ovarian cancer, but that its ability to reflect pathway inhibition and cell proliferation depends on the delay between treatment cessation and PET imaging. The effects of BEZ-235 on ¹⁸F-FDG uptake by tumors were validated by a full range of molecular biology techniques that on day 3 showed the extinction of key proteins implicated in the PI3K/mTOR pathways. These findings were associated with a significant decrease in cell proliferation, but no tumor necrosis or apoptosis was depicted. Cell proliferation and PI3K/mTOR downstream showed complete or almost complete recovery 4 days after treatment cessation, a pattern that was accurately identified by ¹⁸F-FDG SA-PET imaging at a time when no significant change in tumor volume was depicted by SA-CT between the treated and the control animals. At earlier time points after treatment cessation (i.e., 24 and 48 hours following BEZ-235 withdrawal), a lag between PI3K/mTOR pathway recovery and ¹⁸F-FDG uptake that remained unchanged was observed, whereas ¹⁸F-FDG was well correlated to cell proliferation.

The decrease in tumor ¹⁸F-FDG uptake during treatment with AKT inhibition has been previously described by Ma et al. when using rapamycin, an mTOR inhibitor [19], as a consequence of temporarily decreased glycolysis mediated by a disruption of GLUT1 transporter transcription and/or translocation to the plasma membrane, without effect on tumor growth. A translocation of GLUT1 transporters was also reported by Nguyen et al. [20] after 3.5 hours of treatment with LY294002, an AKT inhibitor, in human colon carcinoma cells xenografted into nude mice. In contrast with the study from Ma et al. [19], our results show that the decrease in ¹⁸F-FDG uptake correlates not only to target inhibition but also to a significant decrease in cell proliferation within 3 days of treatment.

Cejka et al. [16] have shown that ¹⁸F-FDG uptake can be used as a surrogate marker for defining the optimal biologic dose of an mTOR inhibitor, and several other preclinical studies have validated the use of ¹⁸F-FDG PET as a marker of response to mTOR or AKT pathway inhibition in various types of human cancers xenografted in immunodeficient mice [15,16,21]. In none of these studies was tumor recovery following drug withdrawal investigated. Therefore, to the best of our knowledge, this study is the first to demonstrate that ¹⁸F-FDG PET may predict cessation of pathway inhibition early after drug withdrawal, although a lag between PI3K/mTOR pathway recovery and metabolic recovery was observed during the first 48 hours after drug withdrawal. Indeed, some tumors exhibited a partial recovery of AKT as early as 24 hours after BEZ-235 drug withdrawal, at a time when both cell proliferation and glucose metabolism were still decreased. pAKT recovery was reached earlier than p4E-BP1 phosphorylation, suggesting a preferential inhibition of mTOR complex 1 (mTORC1) by BEZ-235, compared to mTORC2. Indeed, it has been shown that mTORC1 could exert an inhibitory effect on mTORC2 (through phospho-p70S6K) [22,23], which is able to activate AKT (Ser⁴⁷³ phosphorylation). It could be hypothesized that the maintained inhibition of mTORC1 after BEZ-235 withdrawal, but not of mTORC2, could lead to an mTORC2-mediated AKT phosphorylation, whereas the targets of mTORC1 such as 4E-BP1 remain unphosphorylated.

In our study, treatment with paclitaxel plus BEZ-235 was not superior to BEZ-235 alone. This is in contrast with previous studies showing an increased efficacy of paclitaxel plus a PI3K inhibitor on tumor cell growth and dissemination as compared with either agent alone [13]. In the present study, paclitaxel was given only once, on the basis of what is done in the clinical setting [24]. Repeated administrations could have enhanced the antitumor activity of paclitaxel [25–27]. It is noteworthy that, in our study, different metabolic responses were observed in the paclitaxel and untreated control groups, although no significant difference was noted between the excipients and paclitaxel groups. These results could be related to the action of the excipients by themselves. Indeed, several studies have shown that Cremophor EL, the diluent in which paclitaxel is prepared for clinical use, is a biologically active solvent [28–30].

Finally, our results raise questions when using ¹⁸F-FDG PET for treatment assessment of molecular therapies that either have a short life or target pathway(s) exhibiting recovery early after drug cessation. Indeed, the delay between the last drug intake could lead to inaccurate results if a patient was scanned several days after treatment cessation, at a time when target inhibition is no longer present. This delay could be determined not only by the prescription of the referring oncologist but also by non-adherence to treatment. Adherence, which is the extent to which a person's behavior taking medication corresponds with agreed recommendations from the healthcare provider, is a major and sometimes unrecognized source of variability in the clinical setting [31]. However, adherence has been poorly explored in solid tumors, mainly in the setting of chronic administration [32,33], despite the increased prescription of oral target therapies. To the best of our knowledge, a precise recording of the last drug intake before PET scanning (i.e., the date and if possible, precise hour) is neither routinely done in PET units nor recommended by the European Association of Nuclear Medicine (EANM) [34] or Society of Nuclear Medicine (SNM) [35] guidelines for tumor imaging. This information is unlikely to be obtainable a posteriori, and we feel that it should be recorded in a systematic way, so that its impact on PET results can be assessed. Information regarding this issue can be drawn from a recent study that attempted to determine compliance with PET protocol parameters defined in the imaging charter within a multicenter trial in patients with non-small cell lung cancer (NSCLC) treated with erlotinib [36]. This study demonstrated that more than 90% of patients completed the scheduled PET scans within the expected window (that is 14 and 56 days following treatment initiation) ± 3 days. This suggests that good levels of compliance are achievable if one decides to systematically incorporate information regarding the time passed since the last drug intake into the PET reports.

Conclusion

¹⁸F-FDG SA-PET is a surrogate marker of target inhibition during treatment with BEZ-235 and predicts tumor recovery 4 days after drug withdrawal, but not during the first 48 hours following drug cessation, when a lag between PI3K/mTOR pathway recovery and metabolic recovery is observed. The results of this study suggest that ¹⁸F-FDG PET could be used for therapy monitoring of PI3K/mTOR inhibitors in cisplatin-resistant ovarian cancer. However, they also raise questions regarding the potential impact of the delay between PET imaging and last drug intake on the accuracy of ¹⁸F-FDG PET imaging when evaluating such inhibitors or other molecular therapies that target pathway(s) exhibiting recovery early after drug cessation.